research article about drug discovery

An overview of drug discovery and development

Affiliation.

• 1 Department of biomedical Science, Nazarbayev University School of Medicine, Nur-Sultan 010000, Kazakhstan.
• PMID: 32270704
• DOI: 10.4155/fmc-2019-0307

A new medicine will take an average of 10-15 years and more than US$2 billion before it can reach the pharmacy shelf. Traditionally, drug discovery relied on natural products as the main source of new drug entities, but was later shifted toward high-throughput synthesis and combinatorial chemistry-based development. New technologies such as ultra-high-throughput drug screening and artificial intelligence are being heavily employed to reduce the cost and the time of early drug discovery, but they remain relatively unchanged. However, are there other potentially faster and cheaper means of drug discovery? Is drug repurposing a viable alternative? In this review, we discuss the different means of drug discovery including their advantages and disadvantages.

Keywords: drug repurposing; high throughput; natural sources; small molecule.

Publication types

• Artificial Intelligence
• Drug Development*
• Drug Evaluation, Preclinical

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

• View all journals
• Explore content
• About the journal
• Publish with us
• Sign up for alerts

Drug discovery articles within Nature

Article | 08 May 2024

Accurate structure prediction of biomolecular interactions with AlphaFold 3

• Josh Abramson
• , Jonas Adler
•  &  John M. Jumper

Research Briefing | 08 May 2024

Toad psychedelic points to biological target for antidepressants

A hallucinogenic compound secreted by toads has served as a springboard for research into the therapeutic benefits of psychedelics. The findings suggest that these compounds exert antidepressant effects in part by binding an under-appreciated target in the brain.

News | 08 May 2024

Major AlphaFold upgrade offers boost for drug discovery

Latest version of the AI models how proteins interact with other molecules — but DeepMind restricts access to the tool.

• Ewen Callaway

Article 08 May 2024 | Open Access

Discovery of potent small-molecule inhibitors of lipoprotein(a) formation

Biochemical screening and optimization identify small molecules that inhibit the formation of lipoprotein(a), and these inhibitors reduce the levels of Lp(a) in several animal models, suggesting that they could provide a therapeutic option in humans.

• , Carlos Perez
•  &  Laura F. Michael

News | 02 May 2024

UTIs make life miserable — scientists are finding new ways to tackle them

Researchers are developing vaccines and fresh drug approaches to prevent and treat recurring infections without antibiotics.

• Carissa Wong

Arts Review | 24 April 2024

Las Borinqueñas remembers the forgotten Puerto Rican women who tested the first pill

Clinical trials of the first oral contraceptive recalled in a bold theatre production.

• Mariana Lenharo

News Explainer | 24 April 2024

Plastic pollution: three numbers that support a crackdown

As negotiators haggle over a global treaty to curb plastics pollution, a flood of data outlines how a treaty could make a difference.

• Nicola Jones

Analysis 17 April 2024 | Open Access

Refining the impact of genetic evidence on clinical success

Human genetic evidence increases the success rate of drugs from clinical development to approval but we are still far from reaching peak genetic insights to aid the discovery of targets for more effective drugs.

• Eric Vallabh Minikel
• , Jeffery L. Painter
•  &  Matthew R. Nelson

News Feature | 16 April 2024

Obesity drugs aren’t always forever. What happens when you quit?

Many researchers think that Wegovy and Ozempic should be taken for life, but myriad factors can force people off them.

• McKenzie Prillaman

Article | 10 April 2024

Metabolic rewiring promotes anti-inflammatory effects of glucocorticoids

Glucocorticoids reprogram the mitochondrial metabolism of macrophages, resulting in increased and sustained production of the anti-inflammatory metabolite itaconate and, as a consequence, inhibition of the inflammatory response.

• Jean-Philippe Auger
• , Max Zimmermann
•  &  Gerhard Krönke

Bitter taste receptor activation by cholesterol and an intracellular tastant

Cryo-electron microscopy structures of the type 2 taste receptor TAS2R14 in complex with Ggust and Gi1 identify cholesterol as an orthosteric agonist and the bitter tastant cmpd28.1 as a positive allosteric modulator and agonist.

• Yoojoong Kim
• , Ryan H. Gumpper
•  &  Bryan L. Roth

Comment | 09 April 2024

AI can help to tailor drugs for Africa — but Africans should lead the way

Computational models that require very little data could transform biomedical and drug development research in Africa, as long as infrastructure, trained staff and secure databases are available.

• Gemma Turon
• , Mathew Njoroge
•  &  Kelly Chibale

Article | 08 April 2024

Tumor-selective activity of RAS-GTP inhibition in pancreatic cancer

• Urszula N. Wasko
• , Jingjing Jiang
•  &  Kenneth P. Olive

News | 19 March 2024

‘A landmark moment’: scientists use AI to design antibodies from scratch

Modified protein-design tool could make it easier to tackle challenging drug targets — but AI antibodies are still a long way from reaching the clinic.

Nature Index | 13 March 2024

A spotlight on the stark imbalances of global health research

An expansion of the Nature Index to include more than 60 medical journals has revealed the clear leaders in the field.

News Feature | 28 February 2024

Can non-profits beat antibiotic resistance and soaring drug costs?

Effective, affordable antimicrobial drugs aren’t moneymakers, despite being desperately needed. Can non-profit organizations pick up the slack?

• Maryn McKenna

Article | 28 February 2024

Anti-TIGIT antibody improves PD-L1 blockade through myeloid and T reg cells

A high baseline of intratumoural macrophages and regulatory T cells is associated with better outcomes in patients with non-small cell lung cancer treated with atezolizumab plus tiragolumab, but not with atezolizumab alone.

• Xiangnan Guan
• , Ruozhen Hu
•  &  Namrata S. Patil

Technology Feature | 27 February 2024

How phase separation is revolutionizing biology

Imaging and molecular manipulation reveal how biomolecular condensates form and offer clues to the role of phase separation in health and disease.

• Elie Dolgin

News Explainer | 26 February 2024

‘Breakthrough’ allergy drug: injection protects against severe food reactions

A study suggests that the asthma treatment omalizumab can reduce the risk of dangerous allergic reactions to peanuts and other foods.

• Sara Reardon

News | 21 February 2024

Drug-resistant microbes: ‘brain drain’ is derailing the fight to stop them

A lack of investment is driving researchers who study antimicrobial resistance out of the field, an industry body warns.

• Lilly Tozer

News | 08 February 2024

Mirror-image molecules separated using workhorse of chemistry

The ability to distinguish between left- and right-handed molecules using mass spectrometry could streamline a laborious part of drug discovery.

• Katharine Sanderson

Correspondence | 06 February 2024

Clinical trials: Japan’s opt-out policy raises risks of adverse drug responses

• , Yudai Kaneda
•  &  Tetsuya Tanimoto

Comment | 31 January 2024

Forget lung, breast or prostate cancer: why tumour naming needs to change

The conventional way of classifying metastatic cancers according to their organ of origin is denying people access to drugs that could help them.

• Fabrice André
• , Elie Rassy
•  &  Benjamin Besse

News | 18 January 2024

AlphaFold found thousands of possible psychedelics. Will its predictions help drug discovery?

Researchers have doubted how useful the AI protein-structure tool will be in discovering medicines — now they are learning how to deploy it effectively.

News & Views | 03 January 2024

A new type of antibiotic targets a drug-resistant bacterium

Infections caused by drug-resistant strains of the bacterium Acinetobacter baumannii have been hard to treat in the clinic. A new class of antibiotics has been identified with the potential to tackle these microbes.

• Morgan K. Gugger
•  &  Paul J. Hergenrother

Article 03 January 2024 | Open Access

A novel antibiotic class targeting the lipopolysaccharide transporter

A tethered macrocyclic peptide antibiotic class described here—which shows potent antibacterial activity against carbapenem-resistant Acinetobacter baumannii —blocks the transport of bacterial lipopolysaccharide from the inner membrane to its destination on the outer membrane through inhibition of the LptB 2 FGC complex.

• Claudia Zampaloni
• , Patrizio Mattei
•  &  Kenneth A. Bradley

Correspondence | 02 January 2024

Ethics compliance should not delay biomedical advances

• Francis A. M. Manno

Research Briefing | 20 December 2023

‘Explainable’ AI identifies a new class of antibiotics

An artificial-intelligence graph neural network was trained on experimental data and used to identify chemical substructures that underlie selective antibiotic activity in more than 12 million compounds. This led to the discovery of a class of antibiotics with in vitro and in vivo activity against Gram-positive bacteria, including Staphylococcus aureus .

Article | 20 December 2023

Discovery of a structural class of antibiotics with explainable deep learning

An explainable deep learning model using a chemical substructure-based approach for the exploration of chemical compound libraries identified structural classes of compounds with antibiotic activity and low toxicity.

• , Erica J. Zheng
•  &  James J. Collins

Article 18 December 2023 | Open Access

The energetic and allosteric landscape for KRAS inhibition

Analysis of the effects of more than 26,000 KRAS mutations on abundance and interactions with six other proteins is used to construct an energy landscape of KRAS and identify allosteric drug target sites.

• Chenchun Weng
• , Andre J. Faure
•  &  Ben Lehner

News Feature | 13 December 2023

Weight-loss-drug pioneer: this biochemist finally gained recognition for her work

Svetlana Mojsov led early studies of GLP-1, the hormone behind Wegovy, Ozempic and other blockbusters.

Career Q&A | 20 November 2023

How my MBA helps me keep my donor-funded research centre afloat

Medicinal chemist Susan Winks shares how she keeps money flowing at the drug-discovery centre she helps to manage.

News | 17 November 2023

EU allows use of controversial weedkiller glyphosate for 10 more years

In the wake of a stalemate among member states, the European Commission has decided to approve the herbicide’s continued use.

• Barbara Casassus

Research Briefing | 08 November 2023

Synthetic drug kills fungi but spares kidney cells

Amphotericin B is a clinically vital antifungal drug, but it has high renal toxicity. The compound kills cells by forming sponge-like aggregates on the cell surface that remove molecules called sterols from the cell membrane. High-resolution structures of these sterol sponges guided rational development and synthesis of a new class of antifungals that are better for renal health.

Nature Podcast | 08 November 2023

How to tame a toxic yet life-saving antifungal

Researchers modify drug to prevent kidney damage, and the mystery of the phosphorus at the Milky Way’s edge.

• Benjamin Thompson
•  &  Nick Petrić Howe

Article | 08 November 2023

Tuning sterol extraction kinetics yields a renal-sparing polyene antifungal

A study reports the development of a structural derivative of amphotericin B with broad antifungal activity in mice but without the renal toxicity associated with amphotericin B.

• , Corinne P. Soutar
•  &  Martin D. Burke

Article | 07 November 2023

Recognition of methamphetamine and other amines by trace amine receptor TAAR1

We report on the structures of the TAAR1–G-protein complex when bound to methamphetamine and other amines.

• , You Zheng
•  &  Fei Xu

Career Q&A | 03 November 2023

My path to heading a biotech company

Shadi Farhangrazi describes how she accidentally became a chief executive.

• Raveena Bhambra

Where I Work | 23 October 2023

I tread the delicate line between culture and conservation

Ethno-ecologist Nolwazi Mbongwa investigates the deeply rooted cultural practices of South African traditional healers, and works with conservationists to protect wildlife.

• Linda Nordling

Clinical Briefing | 18 October 2023

An engineered virus shows potential as an immune therapy in glioblastoma

Therapies for aggressive, recurrent glioblastomas are sorely needed but frequently fail in trials. A first-in-human trial of CAN-3110, an engineered herpes simplex virus 1 (HSV1), shows that it is safe and seems to extend survival and stimulate immune responses — particularly in people with antibodies to HSV1.

Editorial | 10 October 2023

AI’s potential to accelerate drug discovery needs a reality check

Companies say the technology will contribute to faster drug development. Independent verification and clinical trials will determine whether this claim holds up.

News | 06 October 2023

Gene therapies for rare diseases are under threat. Scientists hope to save them

As industry steps aside, scientists seek innovative ways to make sure expensive treatments can reach people who need them.

• Heidi Ledford

News Feature | 04 October 2023

Why rings of RNA could be the next blockbuster drug

The commercial success of RNA vaccines for COVID-19 has revved up interest in circular RNAs as the next generation of therapies.

Outlook | 27 September 2023

RSV treatments are here: now the work begins

Efforts to prevent infections and keep vulnerable people out of hospital are beginning to pay off, but deploying these strategies presents new challenges.

• Benjamin Plackett

Antibody therapies set to transform respiratory syncytial virus prevention for babies

Drugs that counter RSV infection can safeguard newborns, offering another mode of protection alongside vaccines.

Better awareness of RSV in older adults is needed to fight a growing burden

Respiratory syncytial virus is usually associated with babies, but the virus can also cause serious disease in older adults and people with chronic medical conditions.

• Rachel Nuwer

For Indigenous infants, RSV prevention is better than a cure

Governments need to put remote communities at the forefront of strategies to prevent the respiratory disease.

• Anna Banerji

News | 22 September 2023

AlphaFold touted as next big thing for drug discovery — but is it?

Questions remain about whether the AI tool for predicting protein structures can really shake up the pharmaceutical industry.

• Carrie Arnold

Comment | 19 September 2023

AI can help to speed up drug discovery — but only if we give it the right data

Artificial-intelligence tools that enable companies to share data about drug candidates while keeping sensitive information safe can unleash the potential of machine learning and cutting-edge lab techniques, for the common good.

• Marissa Mock
• , Suzanne Edavettal
•  &  Alan Russell

News | 14 September 2023

Psychedelic drug MDMA moves closer to US approval following success in PTSD trial

Long-awaited trial data show drug is effective at treating post-traumatic stress disorder in a diversity of people.

Browse broader subjects

• Biological sciences

Browse narrower subjects

• Business strategy in drug development
• Diagnostics
• Drug delivery
• Drug regulation
• Drug safety
• Drug screening
• Medicinal chemistry
• Pharmaceutics
• Pharmacology
• Target identification
• Target validation

Quick links

• Explore articles by subject
• Guide to authors
• Editorial policies

• Open access
• Published: 23 June 2020

Exploring different approaches to improve the success of drug discovery and development projects: a review

• Geoffrey Kabue Kiriiri   ORCID: orcid.org/0000-0001-9814-2258 1 ,
• Peter Mbugua Njogu 2 &
• Alex Njoroge Mwangi 1  

Future Journal of Pharmaceutical Sciences volume  6 , Article number:  27 ( 2020 ) Cite this article

38k Accesses

83 Citations

Metrics details

There has been a significant increase in the cost and timeline of delivering new drugs for clinical use over the last three decades. Despite the increased investments in research infrastructure by pharmaceutical companies and technological advances in the scientific tools available, efforts to increase the number of molecules coming through the drug development pipeline have largely been unfruitful.

A non-systematic review of the current literature was undertaken to enumerate the various strategies employed to improve the success rates in the pharmaceutical research and development. The review covers the exploitation of genomics and proteomics, complementarity of target-based and phenotypic efficacy screening platforms, drug repurposing and repositioning, collaborative research, focusing on underserved therapeutic fields, outsourcing strategy, and pharmaceutical modeling and artificial intelligence. Examples of successful drug discoveries achieved through application of these strategies are highlighted and discussed herein.

Conclusions

Genomics and proteomics have uncovered a wide array of potential drug targets and are facilitative of enhanced scrupulous target identification and validation thus reducing efficacy-related drug attrition. When used complementarily, phenotypic and target-based screening platforms would likely allow serendipitous drug discovery while increasing rationality in drug design. Drug repurposing and repositioning reduces financial risks in drug development accompanied by cost and time savings, while prolonging patent exclusivity hence increased returns on investment to the innovator company. Equally important, collaborative research is facilitative of cross-fertilization and refinement of ideas, while sharing resources and expertise, hence reducing overhead costs in the early stages of drug discovery. Underserved therapeutic fields are niche drug discovery areas that may be used to experiment and launch novel drug targets, while exploiting incentivized benefits afforded by drug regulatory authorities. Outsourcing allows the pharma industries to focus on their core competencies while deriving greater efficiency of specialist contract research organizations. The existing and emerging pharmaceutical modeling and artificial intelligence softwares and tools allow for in silico computation enabling more efficient computer-aided drug design. Careful selection and application of these strategies, singly or in combination, may potentially harness pharmaceutical research and innovation.

Humans have been in a perpetual tug-of-war with diseases since the ancient days. Efforts to contain plagues have been recorded in historical artifacts over the course of our existence. While many remedies had been discovered in the early centuries, it cannot be gainsaid that the twentieth century was a pharmaceutical golden era that brought the bulk of the current repertoire of drugs at our disposal [ 1 ]. The accelerated speed at which drugs were discovered can be attributed in part by a significant leap in the scientific disciplines of biology and organic chemistry. The former facilitated a thorough understanding of the pathophysiological basis of the diseases hence enabled scientists to accurately detail the underlying biochemical derangements leading to the observed disease phenotype. Organic chemistry, on the other hand, was instrumental in the synthetic and/or semi-synthetic derivation of novel drug molecules to address the existing and emerging unmet medical needs [ 2 ]. Serendipitous drug discovery also played a critical role in drug discoveries as exemplified by the manner that penicillins were discovered by the English bacteriologist Alexander Fleming in the year 1928. The golden era of drug discoveries continued for five decades following the discovery of penicillin. The resultant effects of drug discoveries were felt across all spheres of the human race most notably the significant improvements in the quality of life and the prolonged longevity enabling humans to live much longer than ever before [ 3 ].

The number of new drug molecules coming through the drug discovery and development pipeline started dwindling in the 1980s [ 4 ]. A review of the literature reveals that less than one in 10,000 potential drug compounds that begin the drug discovery journey find their way to the clinic [ 5 ]. It is hypothesized that we may have exhausted the low hanging fruits and thereby greater efforts are needed to bring new drugs to the market. The entry bar for new molecules entering clinical utilization has been raised by regulatory agencies which demand that significant advantages over existing therapeutic options be evident for the new drugs to be considered for marketing authorization. These attributes include increased efficacy, higher potency, reduced toxicity, ease of administration, and affordability [ 6 ]. The overreliance on high technology platforms to identify lead compounds coupled with combinatorial chemistry have been associated with yielding highly lipophilic (greasy) molecules that exhibit poor aqueous solubility resulting in poor pharmacokinetics profiles [ 7 ]. These factors have individually and collectively conspired to increase the cost of identifying and developing new drug molecules with the costs currently hovering above US $2.6 billion per molecule. The number of pharma companies with such financial clout and willing to take the financial risk has gradually decreased through mergers and acquisitions over the years [ 8 ].

While most of the diseases that affect humans have satisfactory therapeutic options available, others have limited or ineffective treatment alternatives and continue exerting a heavy burden on countries and societies. Some of the diseases with huge unmet medical need include neoplastic conditions, diabetes, Alzheimer’s disease, immunological disorders, the human immunodeficiency virus-associated acquired immune deficiency syndrome (HIV-AIDs) [ 9 ], neglected tropical diseases (NTDs), and rare diseases [ 10 ]. Absolute curative therapies for these diseases remain elusive providing a compelling necessity for continued search for new drug molecules.

Figure 1 summarizes the drug discovery and development process. Though a highly lucrative and rewarding enterprise, the process of drug discovery and development is a complicated and arduous scientific journey that begins with identification of a disease or disease area with an unmet medical need. The pharmaceutical or biopharmaceutical firm embarks on the pre-discovery phase which entails elucidation of underlying molecular basis of the disease and development of appropriate animal disease models as well as assay platforms. This is followed by identification of putative targets whose chemical modulation may lead to a therapeutic effect. Upon target identification and validation, the drug discovery team embarks on identifying molecules with the desired pharmacological activity starting with primary hit compounds that are systematically modified to enhance the potency, decrease unwanted effects, and improve desirable physicochemical attributes during the hit-to-lead discovery phase. The end product of the drug discovery process is a candidate drug that is taken through pre-clinical studies and later drug development that transforms the molecule to a clinically useful medicinal product whose efficacy, safety, dosing, and tolerability is established through elaborately designed and executed clinical trials [ 11 ].

Generic outline of the drug discovery and development process

Strategies for improved success in the drug discovery and development process

Key approaches.

Several strategic approaches to enhance efficiency in the drug discovery and development process have been proposed, adopted, and exploited to varied extent in the pharmaceutical research and development (R&D) projects. They include exploitation of genomics and proteomics, the complementarity of phenotypic and target-based screening platforms, expanding the use of existing drug molecules through repurposing and repositioning, use of collaborative research, exploring under-served therapeutic areas, outsourcing approach, and pharmaceutical modeling and artificial intelligence.

Exploitation of genomics and proteomics

It is an established fact that majority of diseases have a molecular or genetic etiology [ 12 , 13 ]. Some conditions including sickle cell disease, cystic fibrosis, muscular dystrophy, and Huntington disease are caused by single gene mutations [ 14 ]. Syndromic conditions such as diabetes and cardiovascular diseases have multifactorial causes including multiple gene mutations confounded by environmental and lifestyle factors [ 12 ]. In the concept of drug discovery, genes have therefore been classified as disease genes, disease-modifying genes, and druggable genes [ 15 ]. Disease genes are those whose mutations cause or predispose a person to the development of a given disease [ 16 ]. Disease-modifying genes encode functional proteins whose altered expression is directly linked to the etiology and progression of a given disease. Druggable genes encode proteins that possess recognition domains capable of interacting with drug molecules eliciting a pharmacological response [ 17 ].

In the current era of target-based drug discovery, it is imperative that the target is scrupulously identified and validated to establish its essentiality in the disease phenotype. This prevents downstream attrition with available data indicating that a significant proportion (52%) of drug failure in clinical trials is due to poor efficacy. Figure 2 depicts the various causes of attrition [ 18 , 19 ]. Classical cases of the drugs imatinib and trastuzumab exemplifies the value of careful target identification and validation in enhancing the success of the drug discovery process [ 20 , 21 , 22 ]. While the above were new molecules carefully designed with the knowledge of the underlying genetic mutation, existing drugs may find new applications through repositioning from their approved indications based on information obtained through genomics [ 23 ]. Genomics can be used to identify and validate druggable genes thus expanding the number of targets available for exploration in drug discovery [ 17 , 24 ]. The use of genomics in target validation has expansively widened through advancement in antisense technology, small interfering RNA (siRNA) that mimic the natural RNA interference (RNAi) and transgenic animal models [ 25 ].

Causes of attrition in drug discovery and development

Exploitation of genomics is not restricted to target identification and validation. Rather, recent trends in pharma R&D show that genomics may be employed in the recruitment of study participants for clinical trials with the selection favoring those subjects more likely to benefit from the intervention being trialed. This ensures that the effect of the drug will be evident if the drug is indeed effective against the target disease and absent if ineffective. The outcome so observed would therefore be attributable to the therapeutic intervention and shielded from other confounders. Genomics can also be used as a predictive tool to forecast potential toxicities emanating from a specific molecule [ 22 ]. Not surprising, the discipline of pharmacogenomics where drugs are adapted to meet individual profiles is fast gaining traction among researchers and medical practitioners, and has positively impacted the process of drug discovery and development [ 22 ].

The human genome was fully described in the year 2002, uncovering a vast treasure trove from which a wide array of novel drug targets could be discovered. Nonetheless, the scientific hype that was associated with the genome project has not been followed with solid benefits as less than 500 of the potential 10,000 targets have been utilized according to the repertoire of drugs registered by the United States Food and Drug Administration (US-FDA) [ 1 , 26 ]. These targets are protein molecules including DNA, RNA, G protein-coupled receptors (GPCRs), enzymes, and ion channels. The GPCRs constitute the largest proportion of targets for currently registered molecules [ 27 ]. It is however expected that the genomic revolution will enhance the drug discovery process significantly given the intensive research currently being done in this field [ 28 ].

Proteomics which is a subset of genomics has been widely explored as an avenue of drug discovery [ 29 ]. Proteomics entails identification, characterization, and quantification of cellular proteins with the aim of establishing their role in the disease progression and the underlying potential for chemotherapeutic manipulation [ 25 ]. Proteomics has been applied widely in drug discovery projects for antineoplastics, neurological, cardiovascular, and rare diseases [ 30 ]. Technologies used in proteomics include gel electrophoresis for protein separation and characterization, mass spectrometry (MS) for identification, and yeast hybrid systems to study protein-protein interactions [ 31 ]. These approaches have the potential to identify novel drug targets and their corresponding genes.

Complementarity of phenotypic and target-based screening platforms

Two distinct screening approaches are routinely employed in the efficacy studies, namely phenotypic (whole-cell) screening and target-based (biochemical) screening. Phenotypic screening evaluates the effects of potential drugs on cultured cell lines (in vitro), isolated tissues/organs (ex-vivo), or in whole animals (in vivo) while target-based screening involves testing the molecules on purified target proteins in vitro [ 32 ]. In the first instance, phenotypic screens are primarily aimed at identifying molecules capable of eliciting the desired pharmacological effect without necessarily elucidating the underlying mechanism of action at the molecular level. They are therefore empirically driven as they focus on phenotypic endpoints. Phenotypic drug screening is information-rich, and the therapeutic relevance of the drug is established much earlier in the drug discovery process. The approach is more physiologically relevant as it is conducted in biological systems that simulate the real physiological environment where cognizance that pharmacological effects result from an interplay of many factors is well appreciated [ 33 , 34 ]. It also provides a huge biological space for serendipitous drug discoveries [ 32 , 35 ]. On the contrary, target-based screening is hypothesis-driven, systematic, and rational. Of essence, it requires identification and isolation of a biochemical target whose modulation leads to a desired pharmacological effect. It employs advanced molecular technologies and biological methods that are facilitative of high throughput screening (HTS) platforms [ 36 ].

Whereas phenotypic screening predominated in the decades before 1980, it has largely been de-emphasized as advances in molecular biology, and genomics took root and favored the target-based screening [ 37 ]. The significant decline in the discovery of first-in-class molecules has in part been attributed to an increasing emphasis on the target-based drug discovery approach [ 34 ]. Analysis of data of the drugs registered by the US-FDA reveals that phenotypic drug discovery has yielded more first-in-class molecules than target-based screening [ 38 ]. These findings have been challenged by a study that established that 78 of 113 first-in-class molecules registered between years 1999 and 2013 were discovered using target-based screening approaches [ 39 ]. The key disadvantages of phenotypic assays include low screening capacity when whole animals are used and the impracticality or difficulty of developing appropriate disease models such as for Alzheimer’s disease [ 40 ]. Numerous reports have demonstrated the inaccuracy of animal models as tools in predicting therapeutic efficacy in humans [ 41 ].

Target-based drug discovery has been the predominant approach of screening putative molecules in the last three decades [ 33 , 42 ]. This has majorly been due to advances in cloning technologies that allow isolation of pure proteins that are then used to screen a large library of compounds using HTS. The high screening capacity afforded by this approach has cemented target-based platform as the default drug discovery approach as companies seek a competitive edge to deliver novel molecules to the market [ 36 ]. Target-based drug discovery begins with understanding the pathophysiological basis of the disease and subsequent identification of the errant biochemical pathway that leads to the disease phenotype. The specific protein that is aberrantly expressed is identified, isolated and its role in the disease phenotype validated by modulation using genomic or pharmacological approaches.

Target-based drug discovery, therefore, elucidates the specific mechanism through which potential drugs produce a pharmacological response. While it lags behind the phenotypic drug discovery approach in yielding first-in-class molecules, target-based drug discovery is unrivalled in producing the best-in-class follower molecules [ 38 ]. This is due in part to the rational, hypothesis and systematic approach employed leading to highly selective, potent molecules with better pharmacokinetic and toxicological profiles. Target based-drug discovery has the advantages of being simpler to undertake, enable faster development, and it enables elucidation of the underlying mechanism of action. It also enables the utilization of modern technological advances including computational modeling, molecular biology, combinatorial chemistry, proteomics, and genomics. Conversely, since the approach is based on the modulation of isolated protein targets, the observed effect may have little physiological relevance as there is oversimplification of the physiological environment in which the drug molecules are evaluated [ 43 ].

Pharmacological effects derive from complex interactions in intact physiological systems that are best simulated by phenotypic drug discovery and are therefore more predictive of the ultimate therapeutic effect in human disease compared to target-based approaches. It is however imperative that drugs be rationally designed to afford specificity thus improved toxicological profiles, while also providing well-defined mechanisms of action of the pharmacologically active molecules offering a firm foundation upon which drugs with better pharmacokinetics and pharmacodynamics profiles may be developed. Therefore, complementary application of both approaches will invariably lead to increased efficiency in drug discovery with the phenotypic approach delivering first-in-class molecules with proven efficacy early in the discovery process. Target-based drug discovery will build upon these foundations to deliver superior follower molecules employing the knowledge on the molecular interactions of the active molecules with the target. There has been a resurgence of the use of phenotypic drug discovery process in an effort to reverse the decline in discovery of new molecular entities coming through the drug discovery pipeline [ 34 , 44 , 45 ]. Table 1 gives a summary of the merits and demerits of either approach.

Repurposing and repositioning of existing drug molecules

Drugs that have been developed for a specific therapeutic application may in the course of their clinical use potentially reveal beneficial effects in other therapeutic areas outside the scope of their original indications. These molecules may, therefore, be evaluated for use in the new diseases areas without requiring structural modifications (drug repurposing) [ 46 ]. Alternatively, the drugs may require alteration of the primary molecular structure to accentuate a desirable side activity while diminishing the primary effect (drug repositioning) [ 47 ]. The two approaches have the potential to resuscitate/rescue previously abandoned molecules as well as expanding the therapeutic applications of drugs in current use. Examples of successful applications of drug repurposing and repositioning are given in Table 2 . They include the drug miltefosine which was developed in the 1980s as an antitumor agent but abandoned due to dose-limiting gastrointestinal side effects. The drug was refocused as an antileishmanial drug with significant success [ 49 ]. Other potential applications for its use as an anti-infective agent have been established with the latest, being its use in the treatment of granulomatous amoebic encephalitis [ 50 ].

Sildenafil is another classic example of successful drug repurposing. Although primarily researched for and originally launched into the market for treatment of pulmonary arterial hypertension secondary to patent ductus arteriosus, sildenafil and other phosphodiesterase type 5 inhibitors are best known for their repurposed clinical indication, namely the management of erectile dysfunction [ 51 , 52 ]. Similarly, drug repositioning was efficiently applied in the R&D of antidiabetic sulfonylureas from sulfonamide antibiotics where the hypoglycemic effect was enhanced while diminishing the antibacterial effect through systematic structural modifications [ 53 ]. The key advantage of drug repurposing and repositioning is the faster development time since the pharmacokinetics and toxicological data as well as other pertinent information regarding the molecules are already available with resultant huge economic savings [ 8 , 46 ]. Repurposing remains a viable approach to availing medicines for protozoan diseases and helminthic diseases [ 54 ]. Many experimental drugs that were abandoned due to development issues or efficacy shortfalls could be resuscitated through repurposing/repositioning [ 54 ]. Approaches to repurpose or reposition existing drugs include experimental screening and in silico approaches with the latter utilizing data of existing drugs to identify new molecule with the potential clinical application [ 47 ].

Collaborative research

By its nature, the corporate pharmaceutical industry is highly competitive with each company aspiring to dominate the race to launch new blockbuster molecules. It is an established industry fact that early market entrants reap more than those who launch follower molecules. Pioneer companies are able to establish strong brand recognition as well as patient and physician loyalty before competition enter the market [ 55 ]. Further, early entrants have sufficient time to perfect their product and set the market price. At any given time, the pharma companies are working to discover and develop molecules addressing similar or very closely related drug targets. Given the astronomical funding channeled into pharmaceutical R&D, these duplicated research efforts collectively end up utilizing resources that could better be invested in the R&D of other disease areas with unmet medical needs. A number of collaborative arrangements have been proposed and utilized for greater success of the pharma R&D. These include precompetitive research, pharma-academia collaboration, and public-private partnerships (PPP) models [ 56 ].

The precompetitive research entails collaboration among pharmaceutical companies, biotechnology companies, and the academic drug discovery units that would otherwise compete but are brought together by a common desire to conduct fundamental research that is facilitative of subsequent drug discovery and innovation. In essence, precompetitive research establishes scientific viability of pursuing a given therapeutic pathway prior to initiation of full-throttle drug discovery and development campaign. Some of the areas in which precompetitive research may be practiced include target identification and validation, sharing of compound libraries, and biomarker and assay development. There are numerous benefits deriving from precompetitive collaboration including reduced costs of research as companies share their resources and expertise, greater efficiency as companies focus on their core competencies thus furthering their excellence, and cross-fertilization of scientific ideas [ 57 ]. Precompetitive collaborations are modeled as virtual institutions with scheduled video conferences to monitor and evaluate the progress made. Once the objectives set upon are attained, companies can then venture into separate drug discovery projects [ 58 ]. Renown precompetitive collaborations include the Biomarkers Consortium, Innovative Medicine Initiative and TranSMART [ 59 ]. TransMART is an inter-organizational collaboration including government agencies, academia, and patient advocacy groups that serves as an open data warehouse arising from clinical trials and basic research [ 60 , 61 ]. In recognition of the potential gains that could accrue from precompetitive collaborations, the US-FDA developed guidelines for registration of drugs discovered through collaborative strategies in 2011 [ 62 ].

There exists a strong justification for pharma-academia collaboration. While the pharma industry has the financial muscle to fund drug discovery and development programs, the academia boasts of unrivaled proficiency in the conduct of basic research that delivers lead compounds, animal disease models, and putative drug targets [ 63 ]. The research capacities of academic institutions have been supported by the availability of tools for translational research, HTS, and chemical libraries. Notable pharma-academia collaborations include AstraZeneca-Columbia University, Pfizer-University of California at San Francisco, Monsanto-University of Washington, and the GlaxoSmithKline (GSK)-Harvard University, among others. The most successful partnerships have been in the area of infectious diseases with drug discoveries for malaria and meningitis A being made [ 64 ]. Novo-Nordisk has successfully employed these partnerships to maintain a competitive edge in the field of diabetes and cardiovascular medicine [ 65 ]. Other successful examples include the Scottish Translational Medicine Research Collaboration, the Dundee kinase consortium, structural genetics consortium, Single Nucleotide Polymorphism (SNP) consortium, and the Transcelerate consortium in the USA [ 66 ].

The PPP models play an important role in bringing new drugs to patients. The partnerships involve public institutions, pharmaceutical industries, and the academia. These partnerships help improve the productivity in the pharmaceutical industry while also aiding the development of drug discovery capabilities in academia from publicly funded research [ 67 ]. The World Health Organization-sponsored Special Programme for Research and Training in Tropical Diseases (WHO/TDR) is one of the most notable PPP globally credited with the discovery of several drugs for tropical diseases. They include chlorproguanil-dapsone combination (Lapdap®) with GSK; injectable artemether with Rhone Poulenc Rorer and injectable β-arteether with Artecef for malaria; eflornithine with Marion-Merrill Dow for human African trypanosomiasis; miltefosine with Zentaris; and liposomal amphotericin B with NeXstar for visceral leishmaniasis; ivermectin with Merck for onchocerciasis; and praziquantel with Bayer for schistosomiasis [ 68 , 69 ].

Under-served therapeutic fields

Strategic considerations are vital before a company commits to a drug discovery project. Among the key considerations is the economic viability of a potential drug molecule upon market entry. For sustainable pharma R&D, any drug development candidate must have an acceptable return on investment to ensure the discovery company remains a viable going concern and is able to fund other drugs in the research pipeline. As such majority of the pharmaceutical R&D efforts are inclined to the therapeutic areas with vast economic potential such as oncology, immunotherapy, endocrinology, neurology, and cardiovascular fields where the probability of recouping the huge capital investment is more certain [ 41 ]. Therapeutic areas that offer negligible financial benefits such NTDs and rare diseases do not attract much attention and therefore the opportunities for novel discoveries largely remain unexplored [ 70 ]. Rare diseases are genetic disorders that afflict a small patient population and thus offer little economic promise. The NTDs, on the other hand, are vector-borne diseases that afflict billions of people in resource-poor countries. However, these populations have low purchasing power and as such, the pharma companies may not recoup their investments let alone enjoy profitability [ 71 ].

The NTDs and rare diseases therapeutic areas present potential avenues of discovering novel drug targets that can then be exploited in other more profitable disease areas where they can be of huge economic value. National agencies have incentivized pharma R&D in these areas by providing tax breaks, accelerated reviews, and extended patent exclusivity [ 72 ]. Investments in orphan drugs can serve as a solid platform for new molecules providing a safety net for companies, thus reducing the impact caused by patent expirations on blockbuster medicines [ 73 ]. The rate at which antibiotic resistance is developing outstrips the rate of their development thereby resulting in a decline in the options available for treating infectious diseases. This is also a fertile avenue for pharmaceutical companies to explore [ 74 ].

Outsourcing strategies

The term outsourcing refers to the industrial practice of contracting out services that were previously performed in-house or to access additional capabilities. Outsourcing of certain activities in the drug discovery and development presents an opportunity to enhance the efficiency of the entire process. The outsourcing industry has expanded significantly with the largest growth being registered in China and India where several contract research organizations (CROs) are domiciled supported by cheaper labor, lower land rates, and an increasingly expanding infrastructure [ 75 ]. Some of the activities amenable to outsourcing include target identification and validation, development of disease models, lead discovery and optimization, pre-formulation studies and specific phases or entire clinical trials [ 76 ]. This approach allows pharmaceutical companies to focus on their core competencies while delegating specific activities to the more highly specialized CROs.

Since the contracted firms are specialists in their core areas, outsourcing results in faster development and significant economic savings. Research has indicated that clinical trials that are carried out by CROs have higher success rates compared to those executed by pharmaceutical companies [ 77 ]. Successful outsourced drug discovery and development projects result in cost reduction, increased operational efficiency, and optimization of resource allocation [ 78 ]. Full benefits are only realized when competent partners are selected and the careful implementation of the project followed [ 79 ]. Adequate control measures must be instituted to ensure that the contracted organizations follow the established code of ethics while conducting the trials [ 8 ].

Pharmaceutical modeling and artificial intelligence

Modeling entails the use of in silico simulations to predict diverse attributes of a drug molecule including pharmacokinetics and pharmacodynamics profiles [ 80 ] . Advances in computing power have enabled development of software that allows simulation of the drug-receptor binding processes, a subset of computer-aided drug design (CADD) also referred to as virtual screening, with tremendous benefits to drug discovery efficiency. First, CADD facilitates generation of focused screens that are then validated in vitro. Second, the CADD is well positioned to guide the lead optimization process thus providing valuable information to the medicinal chemistry team aspiring to enhance the lead molecules receptor affinity or optimize drug metabolism and pharmacokinetics (DMPK) properties including absorption, distribution, metabolism, excretion, and the potential for toxicity (ADMET). Third, the CADD facilitates rational drug design either by “growing” starting molecules one functional group at a time (de novo drug design) on the target site or by piecing together fragments into novel molecules (fragment-based drug design) [ 81 ]. Two screening approaches, namely ligand-based virtual screening and target-based virtual screening, have been used in CADD to filter out the compounds that are unlikely to be successful in the development pipeline due to poor physicochemical properties and/or intolerable toxicological profile while identifying those likely to have the activity of interest.

In ligand-based virtual screening, structural features of known compounds are used to construct computer models that are used to predict the properties of other compounds not included in the training data set. The data sets are then used to generate quantitative-structure activity relationship (QSAR) models correlating structural features and the physicochemical properties of a homologous series to the observed biological activity. The chemical structure of known compounds is reduced to a set of molecular descriptors that are used to generate a mathematical model that is used to predict the properties of the test compounds. Molecular descriptors with the highest activity are chosen for the model [ 82 ]. Target-based virtual screening entails computer models that test the docking properties of test compounds against the three-dimensional structure of the target (X-ray crystal structure or homology model) [ 83 , 84 , 85 ]. Each of the test compounds is optimally positioned on the binding site and assigned a score based on the binding affinity. Top scoring compounds are synthesized and tested in vitro [ 86 ]. Application of these models can enhance the efficiency of drug discovery projects by providing focused screens that can have better chances of succeeding downstream. Problematic molecules are also identified earlier in the drug discovery process thus avoiding expensive late-stage failures. Integration of ligand-based and target-based virtual screening yields better results [ 32 , 87 ].

Modeling and simulation have also been employed in various areas of the clinical drug development process [ 88 ]. The modeling process is founded on mathematical polynomials generated from empirical data for real-life patients. Key areas that may be modeled include bio-simulation to inform planning, implementation, and evaluation of clinical trial designs with the goal of optimizing the efficiency, quality, and cost effectiveness of the trials. Pharmacodynamic and pharmacokinetic models are used to predict optimal dosage levels in the various phases of clinical trials as well as in special populations including pediatrics, geriatrics, pregnant women, and others with constrained physiological conditions that will impact on drug disposition. The application of modeling improves the effectiveness of clinical trials with enormous cost and time saving [ 89 , 90 ]. Successful application of computer-based drug design is exemplified by several drugs in clinical use including nelfinavir, imatinib, zanamivir, saquinavir, and norfloxacin [ 91 ].

Artificial intelligence (AI) is increasingly being applied in the drug design and development. This has been possible due to the availability of large chemical and biological databases that are prerequisites for development of accurate predictive models. Scientists contend that AI has the capacity to revolutionize the drug discovery process enabling the screening of billions of potential molecules for hit identification, prioritization of proposed alternatives, and validation of biological targets. It can further guide lead optimization and inform the design and implementation of clinical trials in the latter stages of drug development. Consequently, strategic implementation of AI could enormously supplement the R&D efforts to avail novel, effective, and safe drugs to alleviate human suffering due to unmet clinical needs [ 92 ]. Generative deep learning networks can propose completely new molecules that exhibit the desired physical and biological properties which can be instrumental in the discovery of drugs for complex disease conditions. They may also be in the optimization of existing molecules. AI is also applied in the multi-objective optimization of lead molecules through the application of machine learning allowing identification of compounds that exhibit a healthy balance of the requisite set of physicochemical, biological, and pharmacokinetic characteristics [ 93 ]. Examples of software used in pharmaceutical modeling and AI-guided drug discovery are listed in Table 3 .

The ever-increasing costs of drug discovery projects have not translated into increased efficiency in delivering new medicines. On the contrary, fewer drugs are transiting through the drug development pipeline than ever before. The observed productivity decline is majorly attributable to the overreliance of the industry on high technology platforms, stringent drug registration and approval requirements for new medicines, and the exhaustion of the obvious and easy-to-reach drug targets necessitating exploration of more complex biological systems.

Scientific advancements allow the application of advanced molecular techniques that include genomics and lately proteomics in identification and validation of drug targets. Carefully executed target identification and validation will reduce the attrition rates attributable to poor efficacy that currently accounts for more than 50% of drug failures. The complementarity of phenotypic and target-based drug discovery approaches would enable discovery of first-in-class molecules while also delivering safer, more efficacious and potent best-in-class follower molecules.

Collaborative strategies, such as precompetitive research and public-private partnerships, have positively impacted efficiency in drug discovery. Expansion of research activities into the underserved therapeutic areas covering rare and neglected diseases would offer a safeguard for companies whose blockbuster drugs are teetering on the patent cliff. Advances in computing technologies will also facilitate selection of focused screens with better success rates downstream. Pharmaceutical modeling and AI are expected to continue contributing significantly to improved efficiency in drug discovery and development in the years to come. Carefully executed outsourcing strategies allow companies to focus on their core competencies while delegating other development activities to expertise offered by the CROs, a strategy that accelerates the discovery process while reducing overhead costs.

Availability of data and materials

Data and materials are available upon request.

Abbreviations

Absorption, distribution, metabolism, elimination, and toxicity

• Artificial intelligence

Acquired immune deficiency syndrome

Computer-aided drug design

Contract Research Organization

Drug metabolism and pharmacokinetics

Deoxyribonucleic acid

United States Food and Drug Administration

G protein-coupled receptors

GlaxoSmithKline

High throughput screening

Mass spectrometry

Public-private partnerships

Quantitative-structure activity relationship

Research and development

Ribonucleic acid

RNA interference

Small interfering RNA

Special Programme for Research and Training in Tropical Diseases

World Health Organization

Drews J (2000) Drug discovery: a historical perspective. Science 287:1960–1964 https://doi.org/10.1126/science.287.5460.1960

Article   CAS   PubMed   Google Scholar  

Rotella DP (2016) The critical role of organic chemistry in drug discovery. ACS Chem Neurosci 7:1315–1316 https://doi.org/10.1021/acschemneuro.6b00280

Gaynes R (2017) The discovery of penicillin—new insights after more than 75 years of clinical use. Emerg Infect Dis 23:849–853 https://doi.org/10.3201/eid2305.161556

Article   PubMed Central   Google Scholar  

Pammolli F, Magazzini L, Riccaboni M (2011) The productivity crisis in pharmaceutical R&D. Nat Rev Drug Discov 10:428–438 https://doi.org/10.1038/nrd3405

Petrova E et al (2014) Innovation in the pharmaceutical industry: the process of drug discovery and development. Springer, New York, pp 19–81

Google Scholar  

Naci H, Carter AW, Mossialos E (2015) Why the drug development pipeline is not delivering better medicines. BMJ:h5542 https://doi.org/10.1136/bmj.h5542

Lipinski CA (2000) Drug-like properties and the causes of poor solubility and poor permeability. J Pharmacol Toxicol Methods 44:235–249. https://doi.org/10.1016/S1056-8719 (00)00107-6

Paul SM, Mytelka DS, Dunwiddie CT et al (2010) How to improve R&D productivity: the pharmaceutical industry’s grand challenge. Nat Rev Drug Discov 9:203–214 https://doi.org/10.1038/nrd3078

Cummings JL, Morstorf T, Zhong K (2014) Alzheimer’s disease drug-development pipeline: few candidates, frequent failures. Alzheimers Res Ther 6:37 https://doi.org/10.1186/alzrt269

Article   PubMed   PubMed Central   Google Scholar  

Ridley RG (2002) Medical need, scientific opportunity and the drive for antimalarial drugs. Nature 415:686–693 https://doi.org/10.1038/415686a

Hughes J, Rees S, Kalindjian S, Philpott K (2011) Principles of early drug discovery: principles of early drug discovery. Br J Pharmacol 162:1239–1249 https://doi.org/10.1111/j.1476-5381.2010.01127.x

Article   CAS   PubMed   PubMed Central   Google Scholar  

Wallis G, BASC (1999) The genetic basis of human disease. The Biochemical Society, London

King R, Rotter J, Motulksy A (2002) The genetic basis of common diseases, 2nd ed

WHO | Genes and human disease. In: WHO. https://www.who.int/genomics/public/geneticdiseases/en/index2.html . Accessed 31 Jan 2019

Hopkins AL, Groom CR (2002) The druggable genome. Nat Rev Drug Discov 1:727–730 https://doi.org/10.1038/nrd892

Dixon SJ, Stockwell BR (2009) Identifying druggable disease-modifying gene products. Curr Opin Chem Biol 13:549–555 https://doi.org/10.1016/j.cbpa.2009.08.003

Chris Finan (2017) The druggable genome and support for target identification and validation in drug development. Sci Transl Med 9:eaag1166. https://doi.org/10.1126/scitranslmed.aag1166

Harrison RK (2016) Phase II and phase III failures: 2013–2015. Nat Rev Drug Discov 15:817–818 https://doi.org/10.1038/nrd.2016.184

Hopkins AL (2008) Network pharmacology: the next paradigm in drug discovery. Nat Chem Biol 4:682–690 https://doi.org/10.1038/nchembio.118

Capdeville R, Buchdunger E, Zimmermann J, Matter A (2002) Glivec (STI571, imatinib), a rationally developed, targeted anticancer drug. Nat Rev Drug Discov 1:493–502 https://doi.org/10.1038/nrd839

Cobleigh MA et al (1999) Multinational study of the efficacy and safety of humanized anti-HER2 monoclonal antibody in women who have HER2-overexpressing metastatic breast cancer that has progressed after chemotherapy for metastatic disease. J Clin Oncol 17:2639–2639 https://doi.org/10.1200/JCO.1999.17.9.2639

Garman KS, Nevins JR, Potti A (2007) Genomic strategies for personalized cancer therapy. Hum Mol Genet 16:R226–R232 https://doi.org/10.1093/hmg/ddm184

Lussier YA, Chen JL (2011) The emergence of genome-based drug repositioning. Sci Transl Med 3:96ps35-96ps35. https://doi.org/10.1126/scitranslmed.3001512

Plenge RM, Scolnick EM, Altshuler D (2013) Validating therapeutic targets through human genetics. Nat Rev Drug Discov 12:581–594 https://doi.org/10.1038/nrd4051

Lindsay MA (2003) Target discovery. Nat Rev Drug Discov 2:831–838 https://doi.org/10.1038/nrd1202

Penrod NM, Cowper-Sal-lari R, Moore JH (2011) Systems genetics for drug target discovery. Trends Pharmacol Sci 32:623–630 https://doi.org/10.1016/j.tips.2011.07.002

Bleicher KH, Böhm H-J, Müller K, Alanine AI (2003) Hit and lead generation: beyond high-throughput screening: a guide to drug discovery. Nat Rev Drug Discov 2:369–378 https://doi.org/10.1038/nrd1086

Dimitri Semizarov, Eric Blomme (2008) Genomics in drug discovery and development | Wiley. In: Wiley.com. https://www.wiley.com/en-us/Genomics+in+Drug+Discovery+and+Development-p-9780470096048 . Accessed 19 Mar 2020

Chanda SK, Caldwell JS (2003) Fulfilling the promise: drug discovery in the post-genomic era. Drug Discov Today 8:168–174. https://doi.org/10.1016/S1359-6446 (02)02595-3

Jain KK (2002) Proteomics-based anticancer drug discovery and development. Technol Cancer Res Treat 1:231–236 https://doi.org/10.1177/153303460200100403

Jain KK (2004) Applications of proteomics technologies for drug discovery. In: Hondermarck H (ed) Proteomics: Biomedical and Pharmaceutical Applications. Kluwer Academic Publishers, Dordrecht, pp 201–227

Chapter   Google Scholar  

Moffat JG, Vincent F, Lee JA et al (2017) Opportunities and challenges in phenotypic drug discovery: an industry perspective. Nat Rev Drug Discov 16:531–543 https://doi.org/10.1038/nrd.2017.111

Wagner BK (2016) The resurgence of phenotypic screening in drug discovery and development. Expert Opin Drug Discovery 11:121–125 https://doi.org/10.1517/17460441.2016.1122589

Article   Google Scholar  

Zheng W, Thorne N, McKew JC (2013) Phenotypic screens as a renewed approach for drug discovery. Drug Discov Today 18:1067–1073 https://doi.org/10.1016/j.drudis.2013.07.001

Ban TA (2006) The role of serendipity in drug discovery. Clin Res 8:10

Samsdodd F (2005) Target-based drug discovery: is something wrong? Drug Discov Today 10:139–147. https://doi.org/10.1016/S1359-6446 (04)03316-1

Kotz J (2012) Phenotypic screening, take two. Sci-Bus Exch 5:380–380 https://doi.org/10.1038/scibx.2012.380

Swinney DC (2013) Phenotypic vs. target-based drug discovery for first-in-class medicines. Clin Pharmacol Ther 93:299–301 https://doi.org/10.1038/clpt.2012.236

Eder J, Sedrani R, Wiesmann C (2014) The discovery of first-in-class drugs: origins and evolution. Nat Rev Drug Discov 13:577–587 https://doi.org/10.1038/nrd4336

Khachaturian ZS (2002) Models and modeling systems in Alzheimer disease drug discovery. Alzheimer Dis Assoc Disord 16

Norris, Diana Pankevich, Miriam Davis, Bruce Altevogt (2014) Improving and accelerating therapeutic development for nervous system disorders: workshop summary. National Academies Press (US), Washington (DC)

Swinney DC, Anthony J (2011) How were new medicines discovered? Nat Rev Drug Discov 10:507–519 https://doi.org/10.1038/nrd3480

Croston GE (2017) The utility of target-based discovery. Expert Opin Drug Discovery 12:427–429 https://doi.org/10.1080/17460441.2017.1308351

Comley J (2015) Phenotypic drug discovery: striving towards the highest level of biological relevance. https://www.ddw-online.com/ . Accessed 28 Jan 2019

Sams-Dodd F (2006) Drug discovery: selecting the optimal approach. Drug Discov Today 11:465–472 https://doi.org/10.1016/j.drudis.2006.03.015

Pushpakom S, Iorio F, Eyers PA et al (2018) Drug repurposing: progress, challenges and recommendations. Nat Rev Drug Discov 18:41–58 https://doi.org/10.1038/nrd.2018.168

Article   PubMed   CAS   Google Scholar  

Cha Y, Erez T, Reynolds IJ et al (2018) Drug repurposing from the perspective of pharmaceutical companies: drug repurposing in pharmaceutical companies. Br J Pharmacol 175:168–180 https://doi.org/10.1111/bph.13798

Ashburn TT, Thor KB (2004) Drug repositioning: identifying and developing new uses for existing drugs. Nat Rev Drug Discov 3:673–683 https://doi.org/10.1038/nrd1468

Charlton RL, Rossi-Bergmann B, Denny PW, Steel PG (2018) Repurposing as a strategy for the discovery of new anti-leishmanials: the-state-of-the-art. Parasitology 145:219–236 https://doi.org/10.1017/S0031182017000993

Article   PubMed   Google Scholar  

Sindermann H, Engel J (2006) Development of miltefosine as an oral treatment for leishmaniasis. Trans R Soc Trop Med Hyg 100:S17–S20 https://doi.org/10.1016/j.trstmh.2006.02.010

Board on Health Sciences Policy (2014) Drug repurposing and repositioning: workshop summary. National Academies Press (US)

White JR (2014) A brief history of the development of diabetes medications. Diabetes Spectr 27:82–86 https://doi.org/10.2337/diaspect.27.2.82

Sola D, Rossi L, Schianca GPC et al (2015) State of the art paper sulfonylureas and their use in clinical practice. Arch Med Sci 4:840–848 https://doi.org/10.5114/aoms.2015.53304

Article   CAS   Google Scholar  

Andrews KT, Fisher G, Skinner-Adams TS (2014) Drug repurposing and human parasitic protozoan diseases. Int J Parasitol Drugs Drug Resist 4:95–111 https://doi.org/10.1016/j.ijpddr.2014.02.002

Cha M, Yu F (2018) Pharma’s first-to-market advantage | McKinsey. https://www.mckinsey.com/industries/pharmaceuticals-and-medical-products/our-insights/pharmas-first-to-market-advantage . Accessed 29 Jan 2019

Munos BH, Chin WW (2011) How to revive breakthrough innovation in the pharmaceutical industry. Sci Transl Med 3:89cm16-89cm16. https://doi.org/10.1126/scitranslmed.3002273

Yildirim O, Gottwald M, Schüler P, Michel MC (2016) Opportunities and challenges for drug development: public–private partnerships, adaptive designs and big data. Front Pharmacol 7. https://doi.org/10.3389/fphar.2016.00461

Medicine (US) (2010) Types of pre-competitive collaborations. National Academies Press (US)

Gastfriend E, Lee B (2015) Pre-competitive collaboration in pharma. 19

Athey BD, Braxenthaler M, Haas M, Guo Y. tranSMART: An Open Source and Community-Driven Informatics and Data Sharing Platform for Clinical and Translational Research. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3814495/#__ffn_sectitle .

Schumacher A, et al (2014) A collaborative approach to develop a multi-omics data analytics platform for translational research. https://reader.elsevier.com/reader/sd/pii/S2212066114000350?token=086E8A7B05341EE1C0A558241BF007678C6B310508BD3F97DFD1DEC923B28DE719E0E0DB2B82F6936FC9E544E850DE7F .

Fridlyand J, Simon RM, Walrath JC et al (2013) Considerations for the successful co-development of targeted cancer therapies and companion diagnostics. Nat Rev Drug Discov 12:743–755 https://doi.org/10.1038/nrd4101

Palmer M, Chaguturu R (2017) Academia–pharma partnerships for novel drug discovery: essential or nice to have? Expert Opin Drug Discovery 12:537–540 https://doi.org/10.1080/17460441.2017.1318124

Cui L, Su X (2009) Discovery, mechanisms of action and combination therapy of artemisinin. Expert Rev Anti-Infect Ther 7:999–1013 https://doi.org/10.1586/eri.09.68

Stilz HU, Bregenholt S (2018) Successful pharmaceutical innovation: In: Frølund L, Riedel MF (eds) Strategic Industry-University Partnerships. Elsevier, pp 39–57

Gill D (2014) Re-inventing clinical trials through TransCelerate. Nat Rev Drug Discov 13:787–788 https://doi.org/10.1038/nrd4437

Tralau-Stewart CJ, Wyatt CA, Kleyn DE, Ayad A (2009) Drug discovery: new models for industry–academic partnerships. Drug Discov Today 14:95–101 https://doi.org/10.1016/j.drudis.2008.10.003

Nwaka S, Ridley RG (2003) Virtual drug discovery and development for neglected diseases through public–private partnerships. Nat Rev Drug Discov 2:919–928 https://doi.org/10.1038/nrd1230

Balaña-Fouce R, Pérez Pertejo MY, Domínguez-Asenjo B et al (2019) Walking a tightrope: drug discovery in visceral leishmaniasis. Drug Discov Today 24:1209–1216 https://doi.org/10.1016/j.drudis.2019.03.007

Hong-Bo Weng et al (2018) Innovation in neglected tropical disease drug discovery and development. 7:

Hunter J (2011) Challenges for pharmaceutical industry: new partnerships for sustainable human health. Philos Trans R Soc A Math Phys Eng Sci 369:1817–1825 https://doi.org/10.1098/rsta.2010.0377

Volmar C-H, Wahlestedt C, Brothers SP (2017) Orphan diseases: state of the drug discovery art. Wien Med Wochenschr 167:197–204 https://doi.org/10.1007/s10354-015-0423-0

Sharma A, Jacob A, Tandon M, Kumar D (2010) Orphan drug: development trends and strategies. J Pharm Bioallied Sci 2:290–299 https://doi.org/10.4103/0975-7406.72128

Projan SJ (2003) Why is big pharma getting out of antibacterial drug discovery? Curr Opin Microbiol 6:427–430 https://doi.org/10.1016/j.mib.2003.08.003

Subramaniam S, Dugar S (2012) Outsourcing drug discovery to India and China: from surviving to thriving. Drug Discov Today 17:1055–1058 https://doi.org/10.1016/j.drudis.2012.04.005

Piachaud BS (2002) Outsourcing in the pharmaceutical manufacturing process: an examination of the CRO experience. Technovation 22:81–90. https://doi.org/10.1016/S0166-4972 (01)00081-5

Kaitin KI (2010) Deconstructing the drug development process: the new face of innovation. Clin Pharmacol Ther 87:356–361 https://doi.org/10.1038/clpt.2009.293

Shuchman M (2007) Commercializing clinical trials — risks and benefits of the CRO boom. N Engl J Med 357:1365–1368 https://doi.org/10.1056/NEJMp078176

Klopack TG (2000) Balancing the risks and the benefits. Drug Discov Today 5:157–160. https://doi.org/10.1016/S1359-6446 (00)01469-0

Klebe G (2006) Virtual ligand screening: strategies, perspectives and limitations. Drug Discov Today 11:580–594 https://doi.org/10.1016/j.drudis.2006.05.012

Song CM, Lim SJ, Tong JC (2009) Recent advances in computer-aided drug design. Brief Bioinform 10:579–591 https://doi.org/10.1093/bib/bbp023

Roy K, Kar S, Das RN (2015) Understanding the basics of QSAR for applications in pharmaceutical sciences and risk assessment, 1st ed. Academic Press

Agah S, Faham S, Vaidehi N, Klein-Seetharaman J (2012) Membrane protein structure and dynamics: methods and protocols, 1st ed. Humana Press

Lyne PD (2002) Structure-based virtual screening: an overview. Drug Discov Today 7:1047–1055. https://doi.org/10.1016/S1359-6446 (02)02483-2

Cole JC, Korb O, Olsson TSG, Liebeschuetz J (2011) The basis for target-based virtual screening: protein structures. In: Virtual Screening. pp 87–114

Lindahl ER, Kukol A (2008) Molecular modeling of proteins, 1st ed. Humana Press

Wiggers HJ, Rocha JR, Cheleski J, Montanari CA (2011) Integration of ligand- and target-based virtual screening for the discovery of cruzain inhibitors. Mol Inform 30:565–578 https://doi.org/10.1002/minf.201000146

Stockmann C, Barrett J, Roberts J, Sherwin C (2015) Use of modeling and simulation in the design and conduct of pediatric clinical trials and the optimization of individualized dosing regimens. CPT Pharmacometrics Syst Pharmacol 4:630–640 https://doi.org/10.1002/psp4.12038

Girard P, Cucherat M, Guez D et al (2004) Clinical trial simulation in drug development. Thérapie 59:297–304 https://doi.org/10.2515/therapie:2004057

Office of the Commissioner (2018) About science & research at FDA - how simulation can transform regulatory pathways. Accessed 1 Feb 2019

Prieto-Martínez FD, López-López E, Eurídice Juárez-Mercado K, Medina-Franco JL (2019) Computational drug design methods—current and future perspectives. In: In Silico Drug Design. Elsevier, pp 19–44

Chan HCS, Shan H, Dahoun T et al (2019) Advancing drug discovery via artificial intelligence. Trends Pharmacol Sci 40:592–604 https://doi.org/10.1016/j.tips.2019.06.004

Schneider P, Walters WP, Plowright AT et al (2019) Rethinking drug design in the artificial intelligence era. Nat Rev Drug Discov https://doi.org/10.1038/s41573-019-0050-3

Download references

Acknowledgements

The authors are grateful to the University of Nairobi for providing infrastructural support by way of online resources required for this review.

There was no funding for this manuscript.

Author information

Authors and affiliations.

Department of Pharmaceutics and Pharmacy Practice, School of Pharmacy, University of Nairobi, P.O. Box 19676-00202, Nairobi, Kenya

Geoffrey Kabue Kiriiri & Alex Njoroge Mwangi

Department of Pharmaceutical Chemistry, School of Pharmacy, University of Nairobi, P.O. Box 19676-00202, Nairobi, Kenya

Peter Mbugua Njogu

You can also search for this author in PubMed   Google Scholar

Contributions

GK conceived and wrote the first draft; PN and AN reviewed all drafts and gave additional input to improve the scientific rigor required. The authors read and approved the final draft for publication.

Corresponding author

Correspondence to Geoffrey Kabue Kiriiri .

Ethics declarations

Ethics approval and consent to participate.

Not applicable.

Consent for publication

Competing interests.

There are no competing interests to declare for all authors.

Additional information

Publisher’s note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Cite this article.

Kiriiri, G.K., Njogu, P.M. & Mwangi, A.N. Exploring different approaches to improve the success of drug discovery and development projects: a review. Futur J Pharm Sci 6 , 27 (2020). https://doi.org/10.1186/s43094-020-00047-9

Download citation

Received : 14 November 2019

Accepted : 10 June 2020

Published : 23 June 2020

DOI : https://doi.org/10.1186/s43094-020-00047-9

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Target-based screening
• Repositioning
• Repurposing

• Frontiers in Drug Discovery
• Technologies and Strategies to Enable Drug Discovery
• Research Topics

Drug Discovery and Development Explained: Introductory Notes for the General Public

Total Downloads

Total Views and Downloads

About this Research Topic

Drug discovery and development involve complex processes, highly integrated interdisciplinary research, and collaborations between academic groups and the private sector. It is a long and resource-intensive endeavor characterized by a high attrition rate. Yet new strategies are being explored, aiming at ...

Keywords : Drug discovery, Drug development, General public, Therapeutic agents, Public health data sharing and use

Important Note : All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.

Topic Editors

Topic coordinators, recent articles, submission deadlines, participating journals.

Manuscripts can be submitted to this Research Topic via the following journals:

total views

• Demographics

No records found

total views article views downloads topic views

Top countries

Top referring sites, about frontiers research topics.

With their unique mixes of varied contributions from Original Research to Review Articles, Research Topics unify the most influential researchers, the latest key findings and historical advances in a hot research area! Find out more on how to host your own Frontiers Research Topic or contribute to one as an author.

Artificial Intelligence in Drug Discovery: A Comprehensive Review of Data-driven and Machine Learning Approaches

• Review Paper
• Published: 07 January 2021
• Volume 25 , pages 895–930, ( 2020 )

Cite this article

• Hyunho Kim 1 ,
• Eunyoung Kim 1 ,
• Ingoo Lee 1 ,
• Bongsung Bae 1 ,
• Minsu Park 1 &
• Hojung Nam 1  

4749 Accesses

47 Citations

7 Altmetric

Explore all metrics

As expenditure on drug development increases exponentially, the overall drug discovery process requires a sustainable revolution. Since artificial intelligence (AI) is leading the fourth industrial revolution, AI can be considered as a viable solution for unstable drug research and development. Generally, AI is applied to fields with sufficient data such as computer vision and natural language processing, but there are many efforts to revolutionize the existing drug discovery process by applying AI. This review provides a comprehensive, organized summary of the recent research trends in AI-guided drug discovery process including target identification, hit identification, ADMET prediction, lead optimization, and drug repositioning. The main data sources in each field are also summarized in this review. In addition, an in-depth analysis of the remaining challenges and limitations will be provided, and proposals for promising future directions in each of the aforementioned areas.

Article PDF

Download to read the full article text

Similar content being viewed by others

Review of deep learning: concepts, CNN architectures, challenges, applications, future directions

Artificial intelligence to deep learning: machine intelligence approach for drug discovery

Interpretable scientific discovery with symbolic regression: a review

Avoid common mistakes on your manuscript.

DiMasi, J. A., H. G. Grabowski, and R. W. Hansen (2016) Innovation in the pharmaceutical industry: New estimates of R&D costs. J. Health Econ. 47: 20–33.

Article   Google Scholar  

Paul, S. M., D. S. Mytelka, C. T. Dunwiddie, C. C. Persinger, B. H. Munos, S. R. Lindborg, and A. L. Schacht (2010) How to improve R&D productivity: the pharmaceutical industry's grand challenge. Nat. Rev. Drug Discov. 9: 203–214.

Article   CAS   Google Scholar  

van de Waterbeemd, H. and E. Gifford (2003) ADMET in silico modelling: towards prediction paradise? Nat. Rev. Drug Discov. 2: 192–204.

Mak, K. K. and M. R. Pichika (2019) Artificial intelligence in drug development: present status and future prospects. Drug Discov. Today. 24: 773–780.

Yang, X., Y. Wang, R. Byrne, G. Schneider, and S. Yang (2019) Concepts of artificial intelligence for computer-assisted drug discovery. Chem. Rev. 119: 10520–10594.

Eder, J., R. Sedrani, and C. Wiesmann (2014) The discovery of first-in-class drugs: origins and evolution. Nat. Rev. Drug Discov. 13: 577–587.

Brown, D. (2007) Unfinished business: target-based drug discovery. Drug Discov. Today 12: 1007–1012.

Hsu, Y. H., J. Yao, L. C. Chan, T. J. Wu, J. L. Hsu, Y. F. Fang, Y. Wei, Y. Wu, W. C. Huang, C. L. Liu, Y. C. Chang, M. Y. Wang, C. W. Li, J. Shen, M. K. Chen, A. A. Sahin, A. Sood, G. B. Mills, D. Yu, G. N. Hortobagyi, and M. C. Hung (2014) Definition of PKC-a, CDK6, and MET as therapeutic targets in triple-negative breast cancer. Cancer Res. 74: 4822–4835.

Chen, B. and A. Butte (2016) Leveraging big data to transform target selection and drug discovery. Clin. Pharmacol. Ther. 99: 285–297.

Kodama, K., M. Horikoshi, K. Toda, S. Yamada, K. Hara, J. Irie, M. Sirota, A. A. Morgan, R. Chen, H. Ohtsu, S. Maeda, T. Kadowaki, and A. J. Butte (2012) Expression-based genomewide association study links the receptor CD44 in adipose tissue with type 2 diabetes. Proc. Natl. Acad. Sci. USA. 109: 7049-7054.

Zhu, Z., F. Zhang, H. Hu, A. Bakshi, M. R. Robinson, J. E. Powell, G. W. Montgomery, M. E. Goddard, N. R. Wray, P. M. Visscher, and J. Yang (2016) Integration of summary data from GWAS and eQTL studies predicts complex trait gene targets. Nat. Genet. 48: 481-487.

van Dam, S., U. Võsa, A. van der Graaf, L. Franke, and J. P. de Magalhães (2018) Gene co-expression analysis for functional classification and gene-disease predictions. Brief. Bioinform. 19: 575–592.

Google Scholar  

Petyuk, V. A., R. Chang, M. Ramirez-Restrepo, N. D. Beckmann, M. Y. R. Henrion, P. D. Piehowski, K. Zhu, S. Wang, J. Clarke, M. J. Huentelman, F. Xie, V. Andreev, A. Engel, T. Guettoche, L. Navarro, P. De Jager, J. A. Schneider, C. M. Morris, I. G. McKeith, R. H. Perry, S. Lovestone, R. L. Woltjer, T. G. Beach, L. I. Sue, G. E. Serrano, A. P. Lieberman, R. L. Albin, I. Ferrer, D. C. Mash, C. M. Hulette, J. F. Ervin, E. M. Reiman, J. A. Hardy, D. A. Bennett, E. Schadt, R. D. Smith, and A. J. Myers (2018) The human brainome: network analysis identifies HSPA2 as a novel Alzheimer's disease target. Brain. 141: 2721-2739.

Lee, S., C. Zhang, Z. Liu, M. Klevstig, B. Mukhopadhyay, M. Bergentall, R. Cinar, M. Ståhlman, N. Sikanic, J. K. Park, S. Deshmukh, A. M. Harzandi, T. Kuijpers, M. Grøtli, S. J. Elsässer, B. D. Piening, M. Snyder, U. Smith, J. Nielsen, F. Bäckhed, G. Kunos, M. Uhlen, J. Boren, and A. Mardinoglu (2017) Network analyses identify liver-specific targets for treating liver diseases. Mol. Syst. Biol. 13: 938.

Zou, Q., J. Li, L. Song, X. Zeng, and G. Wang (2016) Similarity computation strategies in the microRNA-disease network: a survey. Brief. Funct. Genomics. 15: 55–64.

Chen, X., D. Xie, L. Wang, Q. Zhao, Z. H. You, and H. Liu (2018) BNPMDA: Bipartite Network Projection for MiRNADisease Association prediction. Bioinformatics. 34: 3178–3186.

Ding, P., J. Luo, C. Liang, Q. Xiao, and B. Cao (2018) Human disease MiRNA inference by combining target information based on heterogeneous manifolds. J. Biomed. Inform. 80: 26–36.

Mohamed, S. K., V. Novácek, and A. Nounu (2020) Discovering protein drug targets using knowledge graph embeddings. Bioinformatics. 36: 603–610.

Richardson, P., I. Griffin, C. Tucker, D. Smith, O. Oechsle, A. Phelan, M. Rawling, E. Savory, and J. Stebbing (2020) Baricitinib as potential treatment for 2019-nCoV acute respiratory disease. Lancet. 395: e30–e31.

Segler, M. H. S., M. Preuss, and M. P. Waller (2018) Planning chemical syntheses with deep neural networks and symbolic AI. Nature. 555: 604–610.

Ferrero, E., I. Dunham, and P. Sanseau (2017) In silico prediction of novel therapeutic targets using gene-disease association data. J. Transl. Med. 15: 182.

Mamoshina, P., M. Volosnikova, I. V. Ozerov, E. Putin, E. Skibina, F. Cortese, and A. Zhavoronkov (2018) Machine learning on human muscle transcriptomic data for biomarker discovery and tissue-specific drug target identification. Front. Genet. 9: 242.

Piñero, J., Á. Bravo, N. Queralt-Rosinach, A. Gutiérrez- Sacristán, J. Deu-Pons, E. Centeno, J. García-García, F. Sanz, and L. I. Furlong (2017) DisGeNET: A comprehensive platform integrating information on human disease-associated genes and variants. Nucleic Acids Res. 45: D833-D839.

Stoeger, T., M. Gerlach, R. I. Morimoto, and L. A. Nunes Amaral (2018) Large-scale investigation of the reasons why potentially important genes are ignored. PLoS Biol. 16: e2006643.

Piñero, J., J. M. Ramírez-Anguita, J. Saüch-Pitarch, F. Ronzano, E. Centeno, F. Sanz, and L. I. Furlong (2020) The DisGeNET knowledge platform for disease genomics: 2019 update. Nucleic Acids Res. 48: D845-D855.

Davis, A. P., C. J. Grondin, R. J. Johnson, D. Sciaky, R. McMorran, J. Wiegers, T. C. Wiegers, and C. J. Mattingly (2019) The Comparative Toxicogenomics Database: update 2019. Nucleic Acids Res. 47: D948-D954.

Vasaikar, S. V., J. Wang, and B. Zhang (2018) LinkedOmics: analyzing multi-omics data within and across 32 cancer types. Nucleic Acids Res. 46: D956–D963.

Carvalho-Silva, D., A. Pierleoni, M. Pignatelli, C. Ong, L. Fumis, N. Karamanis, M. Carmona, A. Faulconbridge, A. Hercules, E. McAuley, A. Miranda, G. Peat, M. Spitzer, J. Barrett, D. G. Hulcoop, E. Papa, G. Koscielny, and I. Dunham (2019) Open Targets Platform: new developments and updates two years on. Nucleic Acids Res. 47: D1056-D1065.

Brown, K. K., M. M. Hann, A. S. Lakdawala, R. Santos, P. J. Thomas, and K. Todd (2018) Approaches to target tractability assessment - a practical perspective. Medchemcomm. 9: 606–613.

Huang, Z., J. Shi, Y. Gao, C. Cui, S. Zhang, J. Li, Y. Zhou, and Q. Cui (2019) HMDD v3.0: a database for experimentally supported human microRNA-disease associations. Nucleic Acids Res. 47: D1013–D1017.

DepMap portal. https://depmap.org/portal/ .

Meyers, R. M., J. G. Bryan, J. M. McFarland, B. A. Weir, A. E. Sizemore, H. Xu, N. V. Dharia, P. G. Montgomery, G. S. Cowley, S. Pantel, A. Goodale, Y. Lee, L. D. Ali, G. Jiang, R. Lubonja, W. F. Harrington, M. Strickland, T. Wu, D. C. Hawes, V. A. Zhivich, M. R. Wyatt, Z. Kalani, J. J. Chang, M. Okamoto, K. Stegmaier, T. R. Golub, J. S. Boehm, F. Vazquez, D. E. Root, W. C. Hahn, and A. Tsherniak (2017) Computational correction of copy number effect improves specificity of CRISPR-Cas9 essentiality screens in cancer cells. Nat. Genet. 49: 1779-1784.

Tsherniak, A., F. Vazquez, P. G. Montgomery, B. A. Weir, G. Kryukov, G. S. Cowley, S. Gill, W. F. Harrington, S. Pantel, J. M. Krill-Burger, R. M. Meyers, L. Ali, A. Goodale, Y. Lee, G. Jiang, J. Hsiao, W. F. J. Gerath, S. Howell, E. Merkel, M. Ghandi, L. A. Garraway, D. E. Root, T. R. Golub, J. S. Boehm, and W. C. Hahn (2017) Defining a cancer dependency map. Cell. 170: 564-576.e16.

Barretina, J., G. Caponigro, N. Stransky, K. Venkatesan, A. A. Margolin, S. Kim, C. J. Wilson, J. Lehár, G. V. Kryukov, D. Sonkin, A. Reddy, M. Liu, L. Murray, M. F. Berger, J. E. Monahan, P. Morais, J. Meltzer, A. Korejwa, J. Jané-Valbuena, F. A. Mapa, J. Thibault, E. Bric-Furlong, P. Raman, A. Shipway, I. H. Engels, J. Cheng, G. K. Yu, J. Yu, P. Aspesi, M. de Silva, K. Jagtap, M. D. Jones, L. Wang, C. Hatton, E. Palescandolo, S. Gupta, S. Mahan, C. Sougnez, R. C. Onofrio, T. Liefeld, L. MacConaill, W. Winckler, M. Reich, N. Li, J. P. Mesirov, S. B. Gabriel, G. Getz, K. Ardlie, V. Chan, V. E. Myer, B. L. Weber, J. Porter, M. Warmuth, P. Finan, J. L. Harris, M. Meyerson, T. R. Golub, M. P. Morrissey, W. R. Sellers, R. Schlegel, and L. A. Garraway (2012) The Cancer Cell Line Encyclopedia enables predictive modelling of anticancer drug sensitivity. Nature. 483: 603-607.

Stransky, N., M. Ghandi, G. V. Kryukov, L. A. Garraway, J. Lehár, M. Liu, D. Sonkin, A. Kauffmann, K. Venkatesan, E. J. Edelman, M. Riester, J. Barretina, G. Caponigro, R. Schlegel, W. R. Sellers, F. Stegmeier, M. Morrissey, A. Amzallag, I. Pruteanu-Malinici, D. A. Haber, S. Ramaswamy, C. H. Benes, M. P. Menden, F. Iorio, M. R. Stratton, U. McDermott, M. J. Garnett, and J. Saez-Rodriguez (2015) Pharmacogenomic agreement between two cancer cell line data sets. Nature. 528: 84-87.

Ghandi, M., F. W. Huang, J. Jané-Valbuena, G. V. Kryukov, C. C. Lo, E. R. McDonald, J. Barretina, E. T. Gelfand, C. M. Bielski, H. Li, K. Hu, A. Y. Andreev-Drakhlin, J. Kim, J. M. Hess, B. J. Haas, F. Aguet, B. A. Weir, M. V. Rothberg, B. R. Paolella, M. S. Lawrence, R. Akbani, Y. Lu, H. L. Tiv, P. C. Gokhale, A. de Weck, A. A. Mansour, C. Oh, J. Shih, K. Hadi, Y. Rosen, J. Bistline, K. Venkatesan, A. Reddy, D. Sonkin, M. Liu, J. Lehar, J. M. Korn, D. A. Porter, M. D. Jones, J. Golji, G. Caponigro, J. E. Taylor, C. M. Dunning, A. L. Creech, A. C. Warren, J. M. McFarland, M. Zamanighomi, A. Kauffmann, N. Stransky, M. Imielinski, Y. E. Maruvka, A. D. Cherniack, A. Tsherniak, F. Vazquez, J. D. Jaffe, A. A. Lane, D. M. Weinstock, C. M. Johannessen, M. P. Morrissey, F. Stegmeier, R. Schlegel, W. C. Hahn, G. Getz, G. B. Mills, J. S. Boehm, T. R. Golub, L. A. Garraway, and W. R. Sellers (2019) Next-generation characterization of the Cancer Cell Line Encyclopedia. Nature. 569: 503-508.

Yu, C., A. M. Mannan, G. M. Yvone, K. N. Ross, Y. L. Zhang, M. A. Marton, B. R. Taylor, A. Crenshaw, J. Z. Gould, P. Tamayo, B. A. Weir, A. Tsherniak, B. Wong, L. A. Garraway, A. F. Shamji, M. A. Palmer, M. A. Foley, W. Winckler, S. L. Schreiber, A. L. Kung, and T. R. Golub (2016) High-throughput identification of genotype-specific cancer vulnerabilities in mixtures of barcoded tumor cell lines. Nat. Biotechnol. 34: 419-423.

Szklarczyk, D., A. L. Gable, D. Lyon, A. Junge, S. Wyder, J. Huerta-Cepas, M. Simonovic, N. T. Doncheva, J. H. Morris, P. Bork, L. J. Jensen, and C. V. Mering (2019) STRING v11: protein-protein association networks with increased coverage, supporting functional discovery in genome-wide experimental datasets. Nucleic Acids Res. 47: D607-D613.

Wang, Y., S. Zhang, F. Li, Y. Zhou, Y. Zhang, Z. Wang, R. Zhang, J. Zhu, Y. Ren, Y. Tan, C. Qin, Y. Li, X. Li, Y. Chen, and F. Zhu (2020) Therapeutic target database 2020: enriched resource for facilitating research and early development of targeted therapeutics. Nucleic Acids Res. 48: D1031-D1041.

Pearson, N., K. Malki, D. Evans, L. Vidler, C. Ruble, J. Scherschel, B. Eastwood, and D. A. Collier (2019) TractaViewer: a genome-wide tool for preliminary assessment of therapeutic target druggability. Bioinformatics. 35: 4509–4510.

Keiser, M. J., V. Setola, J. J. Irwin, C. Laggner, A. I. Abbas, S. J. Hufeisen, N. H. Jensen, M. B. Kuijer, R. C. Matos, T. B. Tran, R. Whaley, R. A. Glennon, J. Hert, K. L. H. Thomas, D. D. Edwards, B. K. Shoichet, and B. L. Roth (2009) Predicting new molecular targets for known drugs. Nature. 462: 175-181.

Morris, G. M., R. Huey, W. Lindstrom, M. F. Sanner, R. K. Belew, D. S. Goodsell, and A. J. Olson (2009) AutoDock4 and AutoDockTools4: Automated docking with selective Receptor flexibility. J. Comput. Chem. 30: 2785–2791.

Trott, O. and A. J. Olson (2010) AutoDock Vina: improving the speed and accuracy of docking with a new scoring function, efficient optimization, and multithreading. J. Comput. Chem. 31: 455–461.

Koes, D. R., M. P. Baumgartner, and C. J. Camacho (2013) Lessons learned in empirical scoring with smina from the CSAR 2011 benchmarking exercise. J. Chem. Inf. Model. 53: 1893–1904.

Ballester, P. J. and J. B. O. Mitchell (2010) A machine learning approach to predicting protein-ligand binding affinity with applications to molecular docking. Bioinformatics. 26: 1169–1175.

Li, L., B. Wang, and S. O. Meroueh (2011) Support vector regression scoring of receptor-ligand complexes for rankordering and virtual screening of chemical libraries. J. Chem. Inf. Model. 51: 2132–2138.

Ragoza, M., J. Hochuli, E. Idrobo, J. Sunseri, and D. R. Koes (2017) Protein-ligand scoring with convolutional neural networks. J. Chem. Inf. Model. 57: 942–957.

Jimenez, J., M. Skalic, G. Martinez-Rosell, and G. De Fabritiis (2018) KDEEP: Protein-ligand absolute binding affinity prediction via 3D-convolutional neural networks. J. Chem. Inf. Model. 58: 287-296.

Imrie, F., A. R. Bradley, M. van der Schaar, and C. M. Deane (2018) Protein family-specific models using deep neural networks and transfer learning improve virtual screening and highlight the need for more data. J. Chem. Inf. Model. 58: 2319-2330.

Stepniewska-Dziubinska, M. M., P. Zielenkiewicz, and P. Siedlecki (2018) Development and evaluation of a deep learning model for protein-ligand binding affinity prediction. Bioinformatics. 34: 3666–3674.

Tian, K., M. Shao, Y. Wang, J. Guan, and S. Zhou (2016) Boosting compound-protein interaction prediction by deep learning. Methods. 110: 64–72.

Feinberg, E. N., D. Sur, Z. Wu, B. E. Husic, H. Mai, Y. Li, S. Sun, J. Yang, B. Ramsundar, and V. S. Pande (2018) PotentialNet for molecular property prediction. ACS Cent. Sci. 4: 1520-1530.

Lim, J., S. Ryu, K. Park, Y. J. Choe, J. Ham, and W. Y. Kim (2019) Predicting drug-target interaction using a novel graph neural network with 3D structure-embedded graph representation. J. Chem. Inf. Model. 59: 3981-3988.

Landrum, G., B. Kelley, P. Tosco, sriniker, gedeck, NadineSchneider, R. Vianello, A. Dalke, AlexanderSavelyev, S. Turk, B. Cole, M. Swain, A. Vaucher, M. Wójcikowski, A. Pahl, JP, strets123, JLVarjo, P. Fuller, DoliathGavid, N. O'Boyle, P. P. Zarrinkar, G. Sforna, M. Nowotka, pzc, J. van Santen, J. H. Jensen, J. Domanski, D. Hall, and P. Avery (2018) rdkit/rdkit: 2018_03_1 (Q1 2018) Release. Zenodo. https://doi.org/10.5281/zenodo.1222070 .

O'Boyle, N. M., M. Banck, C. A. James, C. Morley, T. Vandermeersch, and G. R. Hutchison (2011) Open Babel: An open chemical toolbox. J. Cheminform. 3: 33.

Willighagen, E. L., J. W. Mayfield, J. Alvarsson, A. Berg, L. Carlsson, N. Jeliazkova, S. Kuhn, T. Pluskal, M. Rojas-Cherto, O. Spjuth, G. Torrance, C. T. Evelo, R. Guha, and C. Steinbeck (2017) Erratum to: The Chemistry Development Kit (CDK) v2.0: atom typing, depiction, molecular formulas, and substructure searching. J. Cheminform. 9: 53.

Yap, C. W. (2011) PaDEL-descriptor: an open source software to calculate molecular descriptors and fingerprints. J. Comput. Chem. 32: 1466–1474.

Mauri, A., V. Consonni, M. Pavan, and R. Todeschini (2006) Dragon software: An easy approach to molecular descriptor calculations. Match-Commun. Math. Comput. Chem. 56: 237–248.

Cao, D. S., Y. Z. Liang, J. Yan, G. S. Tan, Q. S. Xu, and S. Liu (2013) PyDPI: freely available python package for chemoinformatics, bioinformatics, and chemogenomics studies. J. Chem. Inf. Model. 53: 3086-3096.

Cao, D. S., N. Xiao, Q. S. Xu, and A. F. Chen (2015) Rcpi: R/ Bioconductor package to generate various descriptors of proteins, compounds and their interactions. Bioinformatics. 31: 279–281.

Moriwaki, H., Y. S. Tian, N. Kawashita, and T. Takagi (2018) Mordred: a molecular descriptor calculator. J. Cheminform. 10: 4.

Burden, F. R. (2001) Quantitative structure-Activity relationship studies using gaussian processes. J. Chem. Inf. Comput Sci. 41: 830–835.

Svetnik, V., A. Liaw, C. Tong, J. C. Culberson, R. P. Sheridan, and B. P. Feuston (2003) Random forest: a classification and regression tool for compound classification and QSAR modeling. J. Chem. Inf. Comput. Sci. 43: 1947–1958.

Ma, J., R. P. Sheridan, A. Liaw, G. E. Dahl, and V. Svetnik (2015) Deep neural nets as a method for quantitative structure- activity relationships. J. Chem. Inf. Model. 55: 263-274.

Xu, Y., J. Ma, A. Liaw, R. P. Sheridan, and V. Svetnik (2017) Demystifying Multitask Deep neural networks for quantitative structure-activity relationships. J. Chem. Inf. Model. 57: 2490–2504.

Ghasemi, F., A. Mehridehnavi, A. Fassihi, and H. Prez-Snchez (2018) Deep neural network in QSAR studies using deep belief network. Appl. Soft Comput. 62: 251–258.

Kato, Y., S. Hamada, and H. Goto (2020) Validation Study of QSAR/DNN models using the competition datasets. Mol. Inf. 39: 1900154.

Lusci, A., G. Pollastri, and P. Baldi (2013) Deep architectures and deep learning in chemoinformatics: the prediction of aqueous solubility for drug-like molecules. J. Chem. Inf. Model. 53: 1563-1575.

Duvenaud, D., D. Maclaurin, J. Aguilera-Iparraguirre, R. Gómez-Bombarelli, T. Hirzel, A. Aspuru-Guzik, and R. P. Adams (2015) Convolutional networks on graphs for learning molecular fingerprints. arXiv. 1509.09292.

Rogers, D. and M. Hahn (2010) Extended-connectivity fingerprints. J. Chem. Inf. Model. 50: 742–754.

Jaeger, S., S. Fulle, and S. Turk (2018) Mol2vec: Unsupervised machine learning approach with chemical intuition. J. Chem. Inf. Model. 58: 27–35.

Chakravarti, S. K. and S. R. M. Alla (2019) Descriptor Free QSAR modeling using deep learning with long short-term memory neural networks. Front. Artif. Intell. 2: 17.

Winter, R., F. Noé, and D. A. Clevert (2019) Learning continuous and data-driven molecular descriptors by translating equivalent chemical representations. Chem. Sci. 10: 1692–1701.

Honda, S., S. Shi, and H. R. Ueda (2019) SMILES transformer: pre-trained molecular fingerprint for low data drug discovery. arXiv. 1911.04738.

Devlin, J., M. W. Chang, K. Lee, and K. Toutanova (2018) BERT: pre-training of deep bidirectional transformers for language understanding. arXiv. 1810.04805.

Altae-Tran, H., B. Ramsundar, A. S. Pappu, and V. Pande (2017) Low data drug discovery with one-shot learning. ACS Cent. Sci. 3: 283–293.

Rohrer, S. G. and K. Baumann (2009) Maximum unbiased validation (MUV) data sets for virtual screening based on PubChem bioactivity data. J. Chem. Inf. Model. 49: 169–184.

Jeon, M., D. Park, J. Lee, H. Jeon, M. Ko, S. Kim, Y. Choi, A. C. Tan, and J. Kang (2019) ReSimNet: drug response similarity prediction using siamese neural networks. Bioinformatics. 35: 5249–5256.

Lamb, J., E. D. Crawford, D. Peck, J. W. Modell, I. C. Blat, M. J. Wrobel, J. Lerner, J. P. Brunet, A. Subramanian, K. N. Ross, M. Reich, H. Hieronymus, G. Wei, S. A. Armstrong, S. J. Haggarty, P. A. Clemons, R. Wei, S. A. Carr, E. S. Lander, and T. R. Golub (2006) The Connectivity Map: using gene-expression signatures to connect small molecules, genes, and disease. Science. 313: 1929-1935.

Park, K., Y. J. Ko, P. Durai, and C. H. Pan (2019) Machine learning-based chemical binding similarity using evolutionary relationships of target genes. Nucleic Acids Res. 47: e128.

Cheng, T., M. Hao, T. Takeda, S. H. Bryant, and Y. Wang (2017) Large-scale prediction of drug-target interaction: a datacentric review. AAPS J. 19: 1264–1275.

Ding, H., I. Takigawa, H. Mamitsuka, and S. Zhu (2014) Similarity-based machine learning methods for predicting drugtarget interactions: a brief review. Brief Bioinform. 15: 734–747.

Bleakley, K. and Y. Yamanishi (2009) Supervised prediction of drug-target interactions using bipartite local models. Bioinformatics. 25: 2397–2403.

Xia, Z., L. Y. Wu, X. Zhou, and S. T. C. Wong (2010) Semisupervised drug-protein interaction prediction from heterogeneous biological spaces. BMC Syst. Biol. 4 Suppl 2: S6.

van Laarhoven, T., S. B. Nabuurs, and E. Marchiori (2011) Gaussian interaction profile kernels for predicting drug-target interaction. Bioinformatics. 27: 3036–3043.

Pahikkala, T., A. Airola, S. Pietila, S. Shakyawar, A. Szwajda, J. Tang, and T. Aittokallio (2015) Toward more realistic drugtarget interaction predictions. Brief. Bioinform. 16: 325-337.

Keum, J. and H. Nam (2017) SELF-BLM: Prediction of drugtarget interactions via self-training SVM. PLoS One. 12: e0171839.

Chen, X., M. X. Liu, and G. Y. Yan (2012) Drug-target interaction prediction by random walk on the heterogeneous network. Mol. Biosyst. 8: 1970–1978.

Luo, Y., X. Zhao, J. Zhou, J. Yang, Y. Zhang, W. Kuang, J. Peng, L. Chen, and J. Zeng (2017) A network integration approach for drug-target interaction prediction and computational drug repositioning from heterogeneous information. Nat. Commun. 8: 573.

Wang, S., H. Cho, C. Zhai, B. Berger, and J. Peng (2015) Exploiting ontology graph for predicting sparsely annotated gene function. Bioinformatics. 31: i357–i364.

Ewing, T., J. C. Baber, and M. Feher (2006) Novel 2D fingerprints for ligand-based virtual screening. J. Chem. Inf. Model. 46: 2423–2431.

Dubchak, I., I. Muchnik, S. R. Holbrook, and S. H. Kim (1995) Prediction of protein folding class using global description of amino acid sequence. Proc. Natl. Acad. Sci. USA. 92: 8700-8704.

Zhang, P., L. Tao, X. Zeng, C. Qin, S. Chen, F. Zhu, Z. Li, Y. Jiang, W. Chen, and Y. Z. Chen (2017) A protein network descriptor server and its use in studying protein, disease, metabolic and drug targeted networks. Brief. Bioinform. 18: 1057-1070.

Yu, H., J. Chen, X. Xu, Y. Li, H. Zhao, Y. Fang, X. Li, W. Zhou, W. Wang, and Y. Wang (2012) A systematic prediction of multiple drug-target interactions from chemical, genomic, and pharmacological data. PLoS One. 7: e37608.

Li, Z. C., M. H. Huang, W. Q. Zhong, Z. Q. Liu, Y. Xie, Z. Dai, and X. Y. Zou (2016) Identification of drug-target interaction from interactome network with ‘guilt-by-association’ principle and topology features. Bioinformatics. 32: 1057–1064.

Lee, I. and H. Nam (2018) Identification of drug-target interaction by a random walk with restart method on an interactome network. BMC Bioinformatics. 19: 208.

Wang, Y. and J. Zeng (2013) Predicting drug-target interactions using restricted Boltzmann machines. Bioinformatics. 29: i126–i134.

Wen, M., Z. Zhang, S. Niu, H. Sha, R. Yang, Y. Yun, and H. Lu (2017) Deep-learning-based drug-target interaction prediction. J. Proteome Res. 16: 1401-1409.

Hu, P. W., K. C. C. Chan, and Z. H. You (2016) Large-scale prediction of drug-target interactions from deep representations. 2016 International Joint Conference on Neural Networks (IJCNN). July 24-29. Vancouver, BC, Canada.

Ozturk, H., A. Ozgur, and E. Ozkirimli (2018) DeepDTA: deep drug-target binding affinity prediction. Bioinformatics. 34: i821–i829.

He, T., M. Heidemeyer, F. Ban, A. Cherkasov, and M. Ester (2017) SimBoost: a read-across approach for predicting drugtarget binding affinities using gradient boosting machines. J. Cheminform. 9: 24.

Tsubaki, M., K. Tomii, and J. Sese (2019) Compound-protein interaction prediction with end-to-end learning of neural networks for graphs and sequences. Bioinformatics. 35: 309–318.

Gonen, M. (2012) Predicting drug-target interactions from chemical and genomic kernels using Bayesian matrix factorization. Bioinformatics. 28: 2304–2310.

Lee, I., J. Keum, and H. Nam (2019) DeepConv-DTI: Prediction of drug-target interactions via deep learning with convolution on protein sequences. PLoS Comput. Biol. 15: e1007129.

Karimi, M., D. Wu, Z. Wang, and Y. Shen (2019) DeepAffinity: interpretable deep learning of compound-protein affinity through unified recurrent and convolutional neural networks. Bioinformatics. 35: 3329-3338.

Shen, C., J. Ding, Z. Wang, D. Cao, X. Ding, and T. Hou (2020) From machine learning to deep learning: Advances in scoring functions for protein-ligand docking. WIREs Comput. Mol. Sci. 10: e1429.

Sieg, J., F. Flachsenberg, and M. Rarey (2019) In need of bias control: evaluating chemical data for machine learning in structure-based virtual screening. J. Chem. Inf. Model. 59: 947-961.

Chen, L., A. Cruz, S. Ramsey, C. J. Dickson, J. S. Duca, V. Hornak, D. R. Koes, and T. Kurtzman (2019) Hidden bias in the DUD-E dataset leads to misleading performance of deep learning in structure-based virtual screening. PLoS One. 14: e0220113.

Hanson, J., K. K. Paliwal, T. Litfin, Y. Yang, and Y. Zhou (2020) Getting to know your neighbor: protein structure prediction comes of age with contextual machine learning. J. Comput. Biol. 27: 796-814.

Shi, Q., W. Chen, S. Huang, Y. Wang, and Z. Xue (2019) Deep learning for mining protein data. Brief. Bioinform. bbz156.

Goodsell, D. S., C. Zardecki, L. Di Costanzo, J. M. Duarte, B. P. Hudson, I. Persikova, J. Segura, C. Shao, M. Voigt, J. D. Westbrook, J. Y. Young, and S. K. Burley (2020) RCSB Protein Data Bank: Enabling biomedical research and drug discovery. Protein Sci. 29: 52-65.

Gola, J., O. Obrezanova, E. Champness, and M. Segall (2006) ADMET property prediction: The state of the art and current challenges. QSAR Comb. Sci. 25: 1172–1180.

Moroy, G., V. Y. Martiny, P. Vayer, B. O. Villoutreix, and M. A. Miteva (2012) Toward in silico structure-based ADMET prediction in drug discovery. Drug Discov. Today. 17: 44–55.

Tian, S., J. Wang, Y. Li, D. Li, L. Xu, and T. Hou (2015) The application of in silico drug-likeness predictions in pharmaceutical research. Adv. Drug Deliv. Rev. 86: 2–10.

Zhao, Y. H., J. Le, M. H. Abraham, A. Hersey, P. J. Eddershaw, C. N. Luscombe, D. Boutina, G. Beck, B. Sherborne, I. Cooper, and J. A. Platts (2001) Evaluation of human intestinal absorption data and subsequent derivation of a quantitative structure-Activity relationship (QSAR) with the Abraham descriptors. J. Pharm. Sci. 90: 749-784.

Ponzoni, I., V. Sebastin-Prez, C. Requena-Triguero, C. Roca, M. J. Martnez, F. Cravero, M. F. Daz, J. A. Pez, R. G. Arrays, J. Adrio, and N. E. Campillo (2017) Hybridizing feature selection and feature learning approaches in QSAR modeling for drug discovery. Sci. Rep. 7: 2403.

Wang, N. N., C. Huang, J. Dong, Z. J. Yao, M. F. Zhu, Z. K. Deng, B. Lv, A. P. Lu, A. F. Chen, and D. S. Cao (2017) Predicting human intestinal absorption with modified random forest approach: a comprehensive evaluation of molecular representation, unbalanced data, and applicability domain issues. RSC Adv. 7: 19007-19018.

Yang, M., J. Chen, L. Xu, X. Shi, X. Zhou, Z. Xi, R. An, and X. Wang (2018) A novel adaptive ensemble classification framework for ADME prediction. RSC Adv. 8: 11661–11683.

Fredlund, L., S. Winiwarter, and C. Hilgendorf (2017) In vitro intrinsic permeability: a transporter-independent measure of Caco-2 cell permeability in drug design and development. Mol. Pharm. 14: 1601-1609.

Patel, R. D., S. P. Kumar, C. N. Patel, S. S. Shankar, H. A. Pandya, and H. A. Solanki (2017) Parallel screening of druglike natural compounds using Caco-2 cell permeability QSAR model with applicability domain, lipophilic ligand efficiency index and shape property: A case study of HIV-1 reverse transcriptase inhibitors. J. Mol. Struct. 1146: 80-95.

Sun, H., K. Nguyen, E. Kerns, Z. Yan, K. R. Yu, P. Shah, A. Jadhav, and X. Xu (2017) Highly predictive and interpretable models for PAMPA permeability. Bioorg. Med. Chem. 25: 1266–1276.

Chi, C. T., M. H. Lee, C. F. Weng, and M. K. Leong (2019) In silico prediction of PAMPA effective permeability using a two-QSAR approach. Int. J. Mol. Sci. 20: 3170.

Lanevskij, K. and R. Didziapetris (2019) Physicochemical QSAR analysis of passive permeability across Caco-2 monolayers. J. Pharm. Sci. 108: 78–86.

Oja, M., S. Sild, and U. Maran (2019) Logistic classification models for pH-permeability profile: predicting permeability classes for the biopharmaceutical classification system. J. Chem. Inf. Model. 59: 2442-2455.

Shin, M., D. Jang, H. Nam, K. H. Lee, and D. Lee (2018) Predicting the absorption potential of chemical compounds through a deep learning approach. IEEE/ACM Trans. Comput. Biol. Bioinform. 15: 432–440.

Wenzel, J., H. Matter, and F. Schmidt (2019) Predictive multitask deep neural network models for ADME-Tox properties: learning from large data sets. J. Chem. Inf. Model. 59: 1253-1268.

Gooch, E. (2004) Medicinal chemistry - an introduction; fundamentals of medicinal chemistry (Gareth Thomas). J. Chem. Educ. 81: 1271.

Kumar, R., A. Sharma, M. H. Siddiqui, and R. K. Tiwari (2017) Prediction of drug-plasma protein binding using artificial intelligence based algorithms. Comb. Chem. High Throughput Screen. 21: 57-64.

Wang, N. N., Z. K. Deng, C. Huang, J. Dong, M. F. Zhu, Z. J. Yao, A. F. Chen, A. P. Lu, Q. Mi, and D. S. Cao (2017) ADME properties evaluation in drug discovery: Prediction of plasma protein binding using NSGA-II combining PLS and consensus modeling. Chemometr. Intell. Lab. Syst. 170: 84-95.

Sun, L., H. Yang, J. Li, T. Wang, W. Li, G. Liu, and Y. Tang (2018) In silico prediction of compounds binding to human plasma proteins by QSAR models. ChemMedChem. 13: 572-581.

Toma, C., D. Gadaleta, A. Roncaglioni, A. Toropov, A. Toropova, M. Marzo, and E. Benfenati (2019) QSAR development for plasma protein binding: influence of the ionization state. Pharm. Res. 36: 28.

Ye, Z., Y. Yang, X. Li, D. Cao, and D. Ouyang (2019) An Integrated transfer learning and multitask learning approach for pharmacokinetic parameter prediction. Mol. Pharm. 16: 533–541.

Prachayasittikul, V., A. Worachartcheewan, A. P. Toropova, A. A. Toropov, N. Schaduangrat, V. Prachayasittikul, and C. Nantasenamat (2017) Large-scale classification of P-glycoprotein inhibitors using SMILES-based descriptors. SAR QSAR Environ. Res. 28: 1-16.

Gonzalo, C. G. and N. García-Pedrajas (2018) Boosted feature selectors: a case study on prediction P-gp inhibitors and substrates. J. Comput. Aided Mol. Des. 32: 1273–1294.

Hinge, V. K., D. Roy, and A. Kovalenko (2019) Prediction of Pglycoprotein inhibitors with machine learning classification models and 3D-RISM-KH theory based solvation energy descriptors. J. Comput. Aided Mol. Des. 33: 965–971.

Shi, T., Y. Yang, S. Huang, L. Chen, Z. Kuang, Y. Heng, and H. Mei (2019) Molecular image-based convolutional neural network for the prediction of ADMET properties. Chemometr. Intell. Lab. Syst. 194: 103853.

Toropov, A. A., A. P. Toropova, M. Beeg, M. Gobbi, and M. Salmona (2017) QSAR model for blood-brain barrier permeation. J. Pharmacol. Toxicol. Methods. 88: 7–18.

Wang, Z., H. Yang, Z. Wu, T. Wang, W. Li, Y. Tang, and G. Liu (2018) In silico prediction of blood-brain barrier permeability of compounds by machine learning and resampling methods. ChemMedChem. 13: 2189-2201.

Yuan, Y., F. Zheng, and C. G. Zhan (2018) Improved prediction of blood-brain barrier permeability through machine learning with combined use of molecular property-based descriptors and fingerprints. AAPS J. 20: 54.

Miao, R., L. Y. Xia, H. H. Chen, H. H. Huang, and Y. Liang (2019) Improved classification of blood-brain-barrier drugs using deep learning. Sci. Rep. 9: 8802.

Hunt, P. A., M. D. Segall, and J. D. Tyzack (2018) WhichP450: a multi-class categorical model to predict the major metabolising CYP450 isoform for a compound. J. Comput. Aided Mol. Des. 32: 537-546.

Tian, S., Y. Djoumbou-Feunang, R. Greiner, and D. S. Wishart (2018) CypReact: A software tool for in silico reactant prediction for human cytochrome P450 enzymes. J. Chem. Inf. Model. 58: 1282-1291.

Shan, X., X. Wang, C. D. Li, Y. Chu, Y. Zhang, Y. Xiong, and D. Q. Wei (2019) Prediction of CYP450 enzyme-substrate selectivity based on the network-based label space division method. J. Chem. Inf. Model. 59: 4577-4586.

Li, X., Y. Xu, L. Lai, and J. Pei (2018) Prediction of human cytochrome P450 inhibition using a multitask deep autoencoder neural network. Mol. Pharm. 15: 4336–4345.

Pang, X., B. Zhang, G. Mu, J. Xia, Q. Xiang, X. Zhao, A. Liu, G. Du, and Y. Cui (2018) Screening of cytochrome P450 3A4 inhibitors via in silico and in vitro approaches. RSC Adv. 8: 34783-34792.

Wu, Z., T. Lei, C. Shen, Z. Wang, D. Cao, and T. Hou (2019) ADMET evaluation in drug discovery. 19. Reliable prediction of human cytochrome P450 inhibition using artificial intelligence approaches. J. Chem. Inf. Model. 59: 4587-4601.

He, S., M. Li, X. Ye, H. Wang, W. Yu, W. He, Y. Wang, and Y. Qiao (2017) Site of metabolism prediction for oxidation reactions mediated by oxidoreductases based on chemical bond. Bioinformatics. 33: 363–372.

Šícho, M., C. De Bruyn Kops, C. Stork, D. Svozil, and J. Kirchmair (2017) FAME 2: simple and effective machine learning model of cytochrome P450 regioselectivity. J. Chem. Inf. Model. 57: 1832-1846.

Finkelmann, A. R., D. D. Goldmann, G. Schneider, and A. H. Goller (2018) MetScore: Site of metabolism prediction beyond cytochrome P450 enzymes. ChemMedChem. 13: 2281–2289.

Cai, Y., H. Yang, W. Li, G. Liu, P. W. Lee, and Y. Tang (2019) Computational prediction of site of metabolism for UGTcatalyzed reactions. J. Chem. Inf. Model. 59: 1085-1095.

Lee, P. W. (2014) Handbook of Metabolic Pathways of Xenobiotics. John Wiley & Sons

Podlewska, S. and R. Kafel (2018) MetStabOn-online platform for metabolic stability predictions. Int. J. Mol. Sci. 19: 1040.

Esaki, T., R. Watanabe, H. Kawashima, R. Ohashi, Y. Natsume-Kitatani, C. Nagao, and K. Mizuguchi (2019) Data curation can improve the prediction accuracy of metabolic intrinsic clearance. Mol. Inform. 38: e1800086.

Liu, K., X. Sun, L. Jia, J. Ma, H. Xing, J. Wu, H. Gao, Y. Sun, F. Boulnois, and J. Fan (2019) Chemi-net: A molecular graph convolutional network for accurate drug property prediction. Int. J. Mol. Sci. 20: 3389.

Zhivkova, Z. D. (2017) Quantitative structure - pharmacokinetic relationships for plasma clearance of basic drugs with consideration of the major elimination pathway. J. Pharm. Pharm. Sci. 20: 135–147.

Wakayama, N., K. Toshimoto, K. Maeda, S. Hotta, T. Ishida, Y. Akiyama, and Y. Sugiyama (2018) In silico prediction of major clearance pathways of drugs among 9 routes with two-step support vector machines. Pharm. Res. 35: 197.

Watanabe, R., R. Ohashi, T. Esaki, H. Kawashima, Y. Natsume-Kitatani, C. Nagao, and K. Mizuguchi (2019) Development of an in silico prediction system of human renal excretion and clearance from chemical structure information incorporating fraction unbound in plasma as a descriptor. Sci. Rep. 9: 18782.

Chen, J., H. Yang, L. Zhu, Z. Wu, W. Li, Y. Tang, and G. Liu (2020) In silico prediction of human renal clearance of compounds using quantitative structure-pharmacokinetic relationship models. Chem. Res. Toxicol. 33: 640-650.

Hong, H., S. Thakkar, M. Chen, and W. Tong (2017) Development of decision forest models for prediction of drug-induced liver injury in humans using a large set of FDA-approved drugs. Sci. Rep. 7: 17311.

Kim, E. and H. Nam (2017) Prediction models for drug-induced hepatotoxicity by using weighted molecular fingerprints. BMC Bioinformatics. 18: 227.

Kotsampasakou, E., F. Montanari, and G. F. Ecker (2017) Predicting drug-induced liver injury: The importance of data curation. Toxicology. 389: 139–145.

Ai, H., W. Chen, L. Zhang, L. Huang, Z. Yin, H. Hu, Q. Zhao, J. Zhao, and H. Liu (2018) Predicting drug-induced liver injury using ensemble learning methods and molecular fingerprints. Toxicol. Sci. 165: 100-107.

Hammann, F., V. Schning, and J. Drewe (2019) Prediction of clinically relevant drug-induced liver injury from structure using machine learning. J. Appl. Toxicol. 39: 412-419.

He, S., T. Ye, R. Wang, C. Zhang, X. Zhang, G. Sun, and X. Sun (2019) An in silico model for predicting drug-induced hepatotoxicity. Int. J. Mol. Sci. 20: 1897.

Williams, D. P., S. E. Lazic, A. J. Foster, E. Semenova, and P. Morgan (2019) Predicting drug-induced liver injury with Bayesian machine learning. Chem. Res. Toxicol. 33: 239-248.

Munawar, S., M. J. Windley, E. G. Tse, M. H. Todd, A. P. Hill, J. I. Vandenberg, and I. Jabeen (2018) Experimentally validated pharmacoinformatics approach to predict hERG inhibition potential of new chemical entities. Front. Pharmacol. 9: 1035.

Siramshetty, V. B., Q. Chen, P. Devarakonda, and R. Preissner (2018) The catch-22 of predicting hERG blockade using publicly accessible bioactivity data. J. Chem. Inf. Model. 58: 1224-1233.

Cai, C., P. Guo, Y. Zhou, J. Zhou, Q. Wang, F. Zhang, J. Fang, and F. Cheng (2019) Deep learning-based prediction of druginduced cardiotoxicity. J. Chem. Inf. Model. 59: 1073-1084.

Konda, L. S. K., S. K. Praba, and R. Kristam (2019) hERG liability classification models using machine learning techniques. Comput. Toxicol. 12: 100089.

Lee, A. A., Q. Yang, A. Bassyouni, C. R. Butler, X. Hou, S. Jenkinson, and D. A. Price (2019) Ligand biological activity predicted by cleaning positive and negative chemical correlations. Proc. Natl. Acad. Sci. USA. 116: 3373-3378.

Lee, H. M., M. S. Yu, S. R. Kazmi, S. Y. Oh, K. H. Rhee, M. A. Bae, B. H. Lee, D. S. Shin, K. S. Oh, H. Ceong, D. Lee, and D. Na (2019) Computational determination of hERG-related cardiotoxicity of drug candidates. BMC Bioinformatics. 20: 250.

Ogura, K., T. Sato, H. Yuki, and T. Honma (2019) Support vector machine model for hERG inhibitory activities based on the integrated hERG database using descriptor selection by NSGA-II. Sci. Rep. 9: 12220.

Zhang, Y., J. Zhao, Y. Wang, Y. Fan, L. Zhu, Y. Yang, X. Chen, T. Lu, Y. Chen, and H. Liu (2019) Prediction of hERG K+ channel blockage using deep neural networks. Chem. Biol. Drug Des. 94: 1973-1985.

Sato, T., H. Yuki, K. Ogura, and T. Honma (2018) Construction of an integrated database for hERG blocking small molecules. PLoS One. 13: e0199348.

Kim, H. and H. Nam (2020) hERG-Att: Self-attention-based deep neural network for predicting hERG blockers. Comput. Biol. Chem. 87: 107286.

Lei, T., F. Chen, H. Liu, H. Sun, Y. Kang, D. Li, Y. Li, and T. Hou (2017) ADMET evaluation in drug discovery. Part 17: development of quantitative and qualitative prediction models for chemical-induced respiratory toxicity. Mol. Pharm. 14: 2407-2421.

Lei, T., H. Sun, Y. Kang, F. Zhu, H. Liu, W. Zhou, Z. Wang, D. Li, Y. Li, and T. Hou (2017) ADMET evaluation in drug discovery. 18. reliable prediction of chemical-induced urinary tract toxicity by boosting machine learning approaches. Mol. Pharm. 14: 3935-3953.

Liu, J., G. Patlewicz, A. J. Williams, R. S. Thomas, and I. Shah (2017) Predicting organ toxicity using in vitro bioactivity data and chemical structure. Chem. Res. Toxicol. 30: 2046–2059.

Xu, Y., J. Pei, and L. Lai (2017) Deep learning based regression and multiclass models for acute oral toxicity prediction with automatic chemical feature extraction. J. Chem. Inf. Model. 57: 2672–2685.

Zhang, H., P. Yu, J. X. Ren, X. B. Li, H. L. Wang, L. Ding, and W. B. Kong (2017) Development of novel prediction model for drug-induced mitochondrial toxicity by using naive Bayes classifier method. Food Chem. Toxicol. 110: 122-129.

Fan, D., H. Yang, F. Li, L. Sun, P. Di, W. Li, Y. Tang, and G. Liu (2018) In silico prediction of chemical genotoxicity using machine learning methods and structural alerts. Toxicol. Res. 7: 211-220.

Jiang, C., H. Yang, P. Di, W. Li, Y. Tang, and G. Liu (2019) In silico prediction of chemical reproductive toxicity using machine learning. J. Appl. Toxicol. 39: 844–854.

Zheng, S., Y. Wang, W. Liu, W. Chang, G. Liang, Y. Xu, and F. Lin (2019) In silico prediction of hemolytic toxicity on the human erythrocytes for small molecules by machine-learning and genetic algorithm. J. Med. Chem. 12: 6499-6512.

Fernandez, M., F. Ban, G. Woo, M. Hsing, T. Yamazaki, E. Leblanc, P. S. Rennie, W. J. Welch, and A. Cherkasov (2018) Toxic colors: the use of deep learning for predicting toxicity of compounds merely from their graphic images. J. Chem. Inf. Model. 58: 1533-1543.

Abbasi, K., A. Poso, J. Ghasemi, M. Amanlou, and A. Masoudi-Nejad (2019) Deep transferable compound representation across domains and tasks for low data drug discovery. J. Chem. Inf. Model. 59: 4528-4539.

Karim, A., A. Mishra, M. A. H. Newton, and A. Sattar (2019) Efficient toxicity prediction via simple features using shallow neural networks and decision trees. ACS Omega. 4: 1874–1888.

Zakharov, A. V., T. Zhao, D. T. Nguyen, T. Peryea, T. Sheils, A. Yasgar, R. Huang, N. Southall, and A. Simeonov (2019) Novel consensus architecture to improve performance of large-scale multitask deep learning QSAR models. J. Chem. Inf. Model. 59: 4613-4624.

Wang, J. and T. Hou (2015) Advances in computationally modeling human oral bioavailability. Adv. Drug Deliv. Rev. 86: 11–16.

Hutter, M. C. (2018) The current limits in virtual screening and property prediction. Future Med. Chem. 10: 1623–1635.

Wu, Z., B. Ramsundar, E. N. Feinberg, J. Gomes, C. Geniesse, A. S. Pappu, K. Leswing, and V. Pande (2018) MoleculeNet: a benchmark for molecular machine learning. Chem. Sci. 9: 513-530.

Merck Molecular Activity Challenge (2012) https://www.kaggle.com/c/MerckActivity .

Winkler, D. A. and T. C. Le (2017) Performance of deep and shallow neural networks, the universal approximation theorem, activity cliffs, and QSAR. Mol. Inform. 36: 1600118.

Ryu, S., Y. Kwon, and W. Y. Kim (2019) A Bayesian graph convolutional network for reliable prediction of molecular properties with uncertainty quantification. Chem. Sci. 10: 8438–8446.

Xiong, Z., D. Wang, X. Liu, F. Zhong, X. Wan, X. Li, Z. Li, X. Luo, K. Chen, H. Jiang, and M. Zheng (2019) Pushing the boundaries of molecular representation for drug discovery with the graph attention mechanism. J. Med. Chem. 63: 8749-8760.

Maggiora, G. M. (2006) On outliers and activity cliffs-Why QSAR often disappoints. J. Chem. Inf. Model. 46: 1535.

Kohonen, P., J. A. Parkkinen, E. L. Willighagen, R. Ceder, K. Wennerberg, S. Kaski, and R. C. Grafstrm (2017) A transcriptomics data-driven gene space accurately predicts liver cytopathology and drug-induced liver injury. Nat. Commun. 8: 15932.

Rueda-Zrate, H. A., I. Imaz-Rosshandler, R. A. Crdenas-Ovando, J. E. Castillo-Fernndez, J. Noguez-Monroy, and C. Rangel-Escareo (2017) A computational toxicogenomics approach identifies a list of highly hepatotoxic compounds from a large microarray database. PLoS One. 12: e0176284.

Su, R., H. Wu, B. Xu, X. Liu, and L. Wei (2019) Developing a multi-dose computational model for drug-induced hepatotoxicity prediction based on toxicogenomics data. IEEE/ACM Trans. Comput. Biol. Bioinform. 16: 1231-1239.

Schneider, G. and U. Fechner (2005) Computer-based de novo design of drug-like molecules. Nat. Rev. Drug Discov. 4: 649–663.

Walters, W. P. (2019) Virtual chemical libraries. J. Med. Chem. 62: 1116–1124.

Reymond, J. L., L. Ruddigkeit, L. Blum, and R. van Deursen (2012) The enumeration of chemical space. WIREs Comput. Mol. Sci. 2: 717-733.

Sanchez-Lengeling, B. and A. Aspuru-Guzik (2018) Inverse molecular design using machine learning: Generative models for matter engineering. Science. 361: 360–365.

Elton, D. C., Z. Boukouvalas, M. D. Fuge, and P. W. Chung (2019) Deep learning for molecular design—a review of the state of the art. Mol. Syst. Des. Eng. 4: 828-849.

Brown, N., M. Fiscato, M. H. S. Segler, and A. C. Vaucher (2019) GuacaMol: Benchmarking models for de novo molecular design. J. Chem. Inf. Model. 59: 1096-1108.

Huc, I. and J. M. Lehn (1997) Virtual combinatorial libraries: dynamic generation of molecular and supramolecular diversity by self-assembly. Proc. Natl. Acad. Sci. USA. 94: 2106–2110.

Lehn, J. M. (1999) Dynamic combinatorial chemistry and virtual combinatorial libraries. Chem. Eur. J. 5: 2455–2463.

Kwon, Y., J. Yoo, Y. S. Choi, W. J. Son, D. Lee, and S. Kang (2019) Efficient learning of non-autoregressive graph variational autoencoders for molecular graph generation. J. Cheminform. 11: 70.

Segler, M. H. S., T. Kogej, C. Tyrchan, and M. P. Waller (2018) Generating focused molecule libraries for drug discovery with recurrent neural networks. ACS Cent. Sci. 4: 120-131.

Gómez-Bombarelli, R., J. N. Wei, D. Duvenaud, J. M. Hernández-Lobato, B. Sánchez-Lengeling, D. Sheberla, J. Aguilera-Iparraguirre, T. D. Hirzel, R. P. Adams, and A. Aspuru-Guzik (2018) Automatic chemical design using a datadriven continuous representation of molecules. ACS Cent. Sci. 4: 268-276.

Kang, S. and K. Cho (2019) Conditional molecular design with deep generative models. J. Chem. Inf. Model. 59: 43–52.

Arús-Pous, J., S. V. Johansson, O. Prykhodko, E. J. Bjerrum, C. Tyrchan, J. L. Reymond, H. Chen, and O. Engkvist (2019) Randomized SMILES strings improve the quality of molecular generative models. J. Cheminform. 11: 71.

Gupta, A., A. T. Müller, B. J. H. Huisman, J. A. Fuchs, P. Schneider, and G. Schneider (2018) Generative recurrent networks for de novo drug design. Mol. Inform. 37: 1700111.

Merk, D., F. Grisoni, L. Friedrich, and G. Schneider (2018) Tuning artificial intelligence on the de novo design of naturalproduct-inspired retinoid X receptor modulators. Commun. Chem. 1: 68.

Zheng, S., X. Yan, Q. Gu, Y. Yang, Y. Du, Y. Lu, and J. Xu (2019) QBMG: quasi-biogenic molecule generator with deep recurrent neural network. J. Cheminform. 11: 5.

Awale, M., F. Sirockin, N. Stiefl, and J. L. Reymond (2019) Drug analogs from fragment-based long short-term memory generative neural networks. J. Chem. Inf. Model. 59: 1347-1356.

Arús-Pous, J., T. Blaschke, S. Ulander, J. L. Reymond, H. Chen, and O. Engkvist (2019) Exploring the GDB-13 chemical space using deep generative models. J. Cheminform. 11: 20.

Pogány, P., N. Arad, S. Genway, and S. D. Pickett (2019) De novo molecule design by translating from reduced graphs to SMILES. J. Chem. Inf. Model. 59: 1136-1146.

Li, Y., L. Zhang, and Z. Liu (2018) Multi-objective de novo drug design with conditional graph generative model. J. Cheminform. 10: 33.

Polykovskiy, D., A. Zhebrak, D. Vetrov, Y. Ivanenkov, V. Aladinskiy, P. Mamoshina, M. Bozdaganyan, A. Aliper, A. Zhavoronkov, and A. Kadurin (2018) Entangled conditional adversarial autoencoder for de novo drug discovery. Mol. Pharm. 15: 4398-4405.

Lim, J., S. Ryu, J. W. Kim, and W. Y. Kim (2018) Molecular generative model based on conditional variational autoencoder for de novo molecular design. J. Cheminform. 10: 31.

Harel, S. and K. Radinsky (2018) Prototype-based compound discovery using deep generative models. Mol. Pharmaceutics. 15: 4406–4416.

Skalic, M., J. Jiménez, D. Sabbadin, and G. De Fabritiis (2019) Shape-based generative modeling for de novo drug design. J. Chem. Inf. Model. 59: 1205-1214.

Lim, J., S. Y. Hwang, S. Moon, S. Kim, and W. Y. Kim (2020) Scaffold-based molecular design with a graph generative model. Chem. Sci. 11: 1153-1164.

Kadurin, A., S. Nikolenko, K. Khrabrov, A. Aliper, and A. Zhavoronkov (2017) druGAN: An advanced generative adversarial autoencoder model for de novo generation of new molecules with desired molecular properties in silico. Mol. Pharmaceutics. 14: 3098-3104.

Blaschke, T., M. Olivecrona, O. Engkvist, J. Bajorath, and H. Chen (2018) Application of generative autoencoder in de novo molecular design. Mol. Inform. 37: 1700123.

Prykhodko, O., S. V. Johansson, P. C. Kotsias, J. Arús-Pous, E. J. Bjerrum, O. Engkvist, and H. Chen (2019) A de novo molecular generation method using latent vector based generative adversarial network. J. Cheminform. 11: 74.

Zhou, Z., S. Kearnes, L. Li, R. N. Zare, and P. Riley (2019) Optimization of molecules via deep reinforcement learning. Sci. Rep. 9: 10752.

Olivecrona, M., T. Blaschke, O. Engkvist, and H. Chen (2017) Molecular de-novo design through deep reinforcement learning. J. Cheminform. 9: 48.

Popova, M., O. Isayev, and A. Tropsha (2018) Deep reinforcement learning for de novo drug design. Sci. Adv. 4: eaap7885.

Putin, E., A. Asadulaev, Y. Ivanenkov, V. Aladinskiy, B. Sanchez-Lengeling, A. Aspuru-Guzik, and A. Zhavoronkov (2018) Reinforced adversarial neural computer for de novo molecular design. J. Chem. Inf. Model. 58: 1194-1204.

Putin, E., A. Asadulaev, Q. Vanhaelen, Y. Ivanenkov, A. V. Aladinskaya, A. Aliper, and A. Zhavoronkov (2018) Adversarial threshold neural computer for molecular de novo design. Mol. Pharmaceutics. 15: 4386-4397.

Liu, X., K. Ye, H. W. T. van Vlijmen, A. P. Ijzerman, and G. J. P. van Westen (2019) An exploration strategy improves the diversity of de novo ligands using deep reinforcement learning: a case for the adenosine A2A receptor. J. Cheminform. 11: 35.

Ståhl, N., G. Falkman, A. Karlsson, G. Mathiason, and J. Boström (2019) Deep reinforcement learning for multiparameter optimization in de novo drug design. J. Chem. Inf. Model. 59: 3166–3176.

Zhavoronkov, A., Y. A. Ivanenkov, A. Aliper, M. S. Veselov, V. A. Aladinskiy, A. V. Aladinskaya, V. A. Terentiev, D. A. Polykovskiy, M. D. Kuznetsov, A. Asadulaev, Y. Volkov, A. Zholus, R. R. Shayakhmetov, A. Zhebrak, L. I. Minaeva, B. A. Zagribelnyy, L. H. Lee, R. Soll, D. Madge, L. Xing, T. Guo, and A. Aspuru-Guzik (2019) Deep learning enables rapid identification of potent DDR1 kinase inhibitors. Nat. Biotechnol. 37: 1038-1040.

Polykovskiy, D., A. Zhebrak, B. Sanchez-Lengeling, S. Golovanov, O. Tatanov, S. Belyaev, R. Kurbanov, A. Artamonov, V. Aladinskiy, M. Veselov, A. Kadurin, S. Johansson, H. Chen, S. Nikolenko, A. Aspuru-Guzik, and A. Zhavoronkov (2018) Molecular Sets (MOSES): A benchmarking platform for molecular generation models. ArXiv. 1811.12823.

Kawai, K., Y. Karuo, A. Tarui, K. Sato, and M. Omote (2020) Effect of structural descriptors on the design of cyclin dependent kinase inhibitors using similarity-based molecular evolution. Mol. Inform. 39: 1900126.

Yoshikawa, N., K. Terayama, M. Sumita, T. Homma, K. Oono, and K. Tsuda (2018) Population-based de novo molecule generation, using grammatical evolution. Chem. Lett. 47: 1431-1434.

Jensen, J. H. (2019) A graph-based genetic algorithm and generative model/Monte Carlo tree search for the exploration of chemical space. Chem. Sci. 10: 3567–3572.

Herring, R. H. and M. R. Eden (2015) Evolutionary algorithm for de novo molecular design with multi-dimensional constraints. Comput. Chem Eng. 83: 267–277.

Rupakheti, C., A. Virshup, W. Yang, and D. N. Beratan (2015) Strategy to discover diverse optimal molecules in the small molecule universe. J. Chem. Inf. Model. 55: 529–537.

Boolell, M., M. J. Allen, S. A. Ballard, S. Gepi-Attee, G. J. Muirhead, A. M. Naylor, I. H. Osterloh, and C. Gingell (1996) Sildenafil: an orally active type 5 cyclic GMP-specific phosphodiesterase inhibitor for the treatment of penile erectile dysfunction. Int. J. Impot Res. 8: 47-52.

Ning, Y. M., J. L. Gulley, P. M. Arlen, S. Woo, S. M. Steinberg, J. J. Wright, H. L. Parnes, J. B. Trepel, M. J. Lee, Y. S. Kim, H. Sun, R. A. Madan, L. Latham, E. Jones, C. C. Chen, W. D. Figg, and W. L. Dahut (2010) Phase II trial of bevacizumab, thalidomide, docetaxel, and prednisone in patients with metastatic castration-resistant prostate cancer. J. Clin. Oncol. 28: 2070-2076.

Singhal, S., J. Mehta, R. Desikan, D. Ayers, P. Roberson, P. Eddlemon, N. Munshi, E. Anaissie, C. Wilson, M. Dhodapkar, J. Zeldis, and B. Barlogie (1999) Antitumor activity of thalidomide in refractory multiple myeloma. N. Engl. J. Med. 341: 1565-1571.

D'Amato, R. J., M. S. Loughnan, E. Flynn, and J. Folkman (1994) Thalidomide is an inhibitor of angiogenesis. Proc. Natl. Acad. Sci. USA. 91: 4082–4085.

Hameed, P. N., K. Verspoor, S. Kusljic, and S. Halgamuge (2018) A two-tiered unsupervised clustering approach for drug repositioning through heterogeneous data integration. BMC Bioinformatics. 19: 129.

Wu, C., R. C. Gudivada, B. J. Aronow, and A. G. Jegga (2013) Computational drug repositioning through heterogeneous network clustering. BMC Syst. Biol. 7: S6.

Blondel, V. D., J. L. Guillaume, R. Lambiotte, and E. Lefebvre (2008) Fast unfolding of communities in large networks. J. Stat. Mech. 2008: P10008.

Nepusz, T., H. Yu, and A. Paccanaro (2012) Detecting overlapping protein complexes in protein-protein interaction networks. Nat. Methods. 9: 471–472.

Sun, P., J. Guo, R. Winnenburg, and J. Baumbach (2017) Drug repurposing by integrated literature mining and drug-genedisease triangulation. Drug Discov. Today. 22: 615–619.

Chen, H. and Z. Zhang (2018) Prediction of drug-disease associations for drug repositioning through drug-miRNAdisease heterogeneous network. IEEE Access. 6: 45281–45287.

Martinez, V., C. Navarro, C. Cano, W. Fajardo, and A. Blanco (2015) DrugNet: network-based drug-disease prioritization by integrating heterogeneous data. Artif. Intell. Med. 63: 41–49.

Martinez, V., C. Cano, and A. Blanco (2014) ProphNet: a generic prioritization method through propagation of information. BMC Bioinformatics. 15: S5.

Luo, H., J. Wang, M. Li, J. Luo, X. Peng, F. X. Wu, and Y. Pan (2016) Drug repositioning based on comprehensive similarity measures and Bi-Random walk algorithm. Bioinformatics. 32: 2664–2671.

Luo, H., M. Li, S. Wang, Q. Liu, Y. Li, and J. Wang (2018) Computational drug repositioning using low-rank matrix approximation and randomized algorithms. Bioinformatics. 34: 1904–1912.

Yan, C. K., W. X. Wang, G. Zhang, J. L. Wang, and A. Patel (2019) BiRWDDA: A novel drug repositioning method based on multisimilarity fusion. J. Comput. Biol. 26: 1230–1242.

Gottlieb, A., G. Y. Stein, E. Ruppin, and R. Sharan (2011) PREDICT: A method for inferring novel drug indications with application to personalized medicine. Mol. Syst. Biol. 7: 496.

Napolitano, F., Y. Zhao, V. M. Moreira, R. Tagliaferri, J. Kere, M. D'Amato, and D. Greco (2013) Drug repositioning: A machinelearning approach through data integration. J. Cheminform. 5: 30.

Wang, Y., S. Chen, N. Deng, and Y. Wang (2013) Drug repositioning by kernel-based integration of molecular structure, molecular activity, and phenotype data. PLoS One. 8: e78518.

Kim, E., A. S. Choi, and H. Nam (2019) Drug repositioning of herbal compounds via a machine-learning approach. BMC Bioinformatics. 20: 247.

Zhang, W., X. Yue, F. Huang, R. Liu, Y. Chen, and C. Ruan (2018) Predicting drug-disease associations and their therapeutic function based on the drug-disease association bipartite network. Methods. 145: 51–59.

Le, D. H. and D. Nguyen-Ngoc (2018) Drug repositioning by integrating known disease-gene and drug-target associations in a semi-supervised learning model. Acta Biotheor. 66: 315–331.

Xuan, P., Y. Cao, T. Zhang, X. Wang, S. Pan, and T. Shen (2019) Drug repositioning through integration of prior knowledge and projections of drugs and diseases. Bioinformatics. 35: 4108–4119.

Wei, X., Y. Zhang, Y. Huang, and Y. Fang (2019) Predicting drug-disease associations by network embedding and biomedical data integration. Data Technol. Appl. 53: 217–229.

Moridi, M., M. Ghadirinia, A. Sharifi-Zarchi, and F. Zare-Mirakabad (2019) The assessment of efficient representation of drug features using deep learning for drug repositioning. BMC Bioinformatics. 20: 577.

Abdolhosseini, F., B. Azarkhalili, A. Maazallahi, A. Kamal, S. A. Motahari, A. Sharifi-Zarchi, and H. Chitsaz (2019) Cell identity codes: understanding cell identity from gene expression profiles using deep neural networks. Sci. Rep. 9: 2342.

Asgari, E. and M. R. K. Mofrad (2015) Continuous distributed representation of biological sequences for deep proteomics and genomics. PLoS One. 10: e0141287.

Donner, Y., S. Kazmierczak, and K. Fortney (2018) Drug Repurposing using deep embeddings of gene expression profiles. Mol. Pharm. 15: 4314–4325.

Stathias, V., J. Turner, A. Koleti, D. Vidovic, D. Cooper, M. Fazel-Najafabadi, M. Pilarczyk, R. Terryn, C. Chung, A. Umeano, D. J. B. Clarke, A. Lachmann, J. E. Evangelista, A. Ma'ayan, M. Medvedovic, and S. C. Schurer (2020) LINCS Data Portal 2.0: next generation access point for perturbationresponse signatures. Nucleic Acids Res. 48: D431-D439.

You, J., R. D. McLeod, and P. Hu (2019) Predicting drug-target interaction network using deep learning model. Comput. Biol. Chem. 80: 90-101.

Aliper, A., S. Plis, A. Artemov, A. Ulloa, P. Mamoshina, and A. Zhavoronkov (2016) Deep learning applications for predicting pharmacological properties of drugs and drug repurposing using transcriptomic data. Mol. Pharm. 13: 2524-2530.

Zeng, X., S. Zhu, X. Liu, Y. Zhou, R. Nussinov, and F. Cheng (2019) deepDR: a network-based deep learning approach to in silico drug repositioning. Bioinformatics. 35: 5191–5198.

Xuan, P., L. Zhao, T. Zhang, Y. Ye, and Y. Zhang (2019) Inferring drug-related diseases based on convolutional neural network and gated recurrent unit. Molecules. 24: 2712.

Masoudi-Sobhanzadeh, Y., Y. Omidi, M. Amanlou, and A. Masoudi-Nejad (2019) Drug databases and their contributions to drug repurposing. Genomics. 112: 1087–1095.

Cheng, F. (2019) In silico oncology drug repositioning and polypharmacology. Methods Mol. Biol. 1878: 243–261.

March-Vila, E., L. Pinzi, N. Sturm, A. Tinivella, O. Engkvist, H. Chen, and G. Rastelli (2017) On the integration of in silico drug design methods for drug repurposing. Front. Pharmacol. 8: 298.

Fleuren, W. W. M. and W. Alkema (2015) Application of text mining in the biomedical domain. Methods. 74: 97–106.

Nugent, T., V. Plachouras, and J. L. Leidner (2016) Computational drug repositioning based on side-effects mined from social media. PeerJ. Computer Science. 2: e46.

Rastegar-Mojarad, M., R. K. Elayavilli, D. Li, R. Prasad, and H. Liu (2015) A new method for prioritizing drug repositioning candidates extracted by literature-based discovery. Proceedings of 2015 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2015. November 9-12. Washington, DC, USA.

Su, E. W. and T. M. Sanger (2017) Systematic drug repositioning through mining adverse event data in ClinicalTrials.gov. PeerJ. 5: e3154.

Park, K. (2019) A review of computational drug repurposing. Transl. Clin. Pharmacol. 27: 59–63.

RDKit. http://www.rdkit.org/ .

Douguet, D. (2018) Data sets representative of the structures and experimental properties of FDA-approved drugs. ACS Med. Chem. Lett. 9: 204–209.

Kim, S., P. A. Thiessen, E. E. Bolton, J. Chen, G. Fu, A. Gindulyte, L. Han, J. He, S. He, B. A. Shoemaker, J. Wang, B. Yu, J. Zhang, and S. H. Bryant (2016) PubChem substance and compound databases. Nucleic Acids Res. 44: D1202-D1213.

Williams, A. J. (2008) Internet-based tools for communication and collaboration in chemistry. Drug Discovery Today. 13: 502–506.

Ursu, O., J. Holmes, C. G. Bologa, J. J. Yang, S. L. Mathias, V. Stathias, D. T. Nguyen, S. Schurer, and T. Oprea (2019) DrugCentral 2018: an update. Nucleic Acids Res. 47: D963-D970.

Ursu, O., J. Holmes, J. Knockel, C. G. Bologa, J. J. Yang, S. L. Mathias, S. J. Nelson, and T. I. Oprea (2017) DrugCentral: online drug compendium. Nucleic Acids Res. 45: D932-D939.

DailyMed. https://dailymed.nlm.nih.gov/dailymed/ .

Kuhn, M., I. Letunic, L. J. Jensen, and P. Bork (2016) The SIDER database of drugs and side effects. Nucleic Acids Res. 44: D1075–D1079.

Tatonetti, N. P., P. P. Ye, R. Daneshjou, and R. B. Altman (2012) Data-driven prediction of drug effects and interactions. Sci. Transl. Med. 4: 125ra31.

Fang, H., Z. Su, Y. Wang, A. Miller, Z. Liu, P. C. Howard, W. Tong, and S. M. Lin (2014) Exploring the FDA adverse event reporting system to generate hypotheses for monitoring of disease characteristics. Clin. Pharmacol. Ther. 95: 496-498.

Cai, M. C., Q. Xu, Y. J. Pan, W. Pan, N. Ji, Y. B. Li, H. J. Jin, K. Liu, and Z. L. Ji (2015) ADReCS: An ontology database for aiding standardization and hierarchical classification of adverse drug reaction terms. Nucleic Acids Res. 43: D907-D913.

Subramanian, A., R. Narayan, S. M. Corsello, D. D. Peck, T. E. Natoli, X. Lu, J. Gould, J. F. Davis, A. A. Tubelli, J. K. Asiedu, D. L. Lahr, J. E. Hirschman, Z. Liu, M. Donahue, B. Julian, M. Khan, D. Wadden, I. C. Smith, D. Lam, A. Liberzon, C. Toder, M. Bagul, M. Orzechowski, O. M. Enache, F. Piccioni, S. A. Johnson, N. J. Lyons, A. H. Berger, A. F. Shamji, A. N. Brooks, A. Vrcic, C. Flynn, J. Rosains, D. Y. Takeda, R. Hu, D. Davison, J. Lamb, K. Ardlie, L. Hogstrom, P. Greenside, N. S. Gray, P. A. Clemons, S. Silver, X. Wu, W. N. Zhao, W. Read-Button, X. Wu, S. J. Haggarty, L. V. Ronco, J. S. Boehm, S. L. Schreiber, J. G. Doench, J. A. Bittker, D. E. Root, B. Wong, and T. R. Golub (2017) A next generation Connectivity Map: L1000 platform and the first 1,000,000 profiles. Cell. 171: 1437-1452.e17.

Barrett, T., D. B. Troup, S. E. Wilhite, P. Ledoux, D. Rudnev, C. Evangelista, I. F. Kim, A. Soboleva, M. Tomashevsky, and R. Edgar (2007) NCBI GEO: Mining tens of millions of expression profiles - Database and tools update. Nucleic Acids Res. 35: D760-D765.

Barrett, T., T. O. Suzek, D. B. Troup, S. E. Wilhite, W. C. Ngau, P. Ledoux, D. Rudnev, A. E. Lash, W. Fujibuchi, and R. Edgar (2005) NCBI GEO: Mining millions of expression profiles -Database and tools. Nucleic Acids Res. 33: D562-D566.

Parkinson, H., M. Kapushesky, M. Shojatalab, N. Abeygunawardena, R. Coulson, A. Farne, E. Holloway, N. Kolesnykov, P. Lilja, M. Lukk, R. Mani, T. Rayner, A. Sharma, E. William, U. Sarkans, and A. Brazma (2007) ArrayExpress - A public database of microarray experiments and gene expression profiles. Nucleic Acids Res. 35: D747-750.

Yang, W., J. Soares, P. Greninger, E. J. Edelman, H. Lightfoot, S. Forbes, N. Bindal, D. Beare, J. A. Smith, I. R. Thompson, S. Ramaswamy, P. A. Futreal, D. A. Haber, M. R. Stratton, C. Benes, U. McDermott, and M. J. Garnett (2013) Genomics of Drug Sensitivity in Cancer (GDSC): A resource for therapeutic biomarker discovery in cancer cells. Nucleic Acids Res. 41: D955-D961.

Bodenreider, O. (2004) The Unified Medical Language System (UMLS): integrating biomedical terminology. Nucleic Acids Res. 32: D267–D270.

Rogers, F. B. (1963) Medical subject headings. Bull. Med. Libr. Assoc. 51: 114–116.

Piñero, J., N. Queralt-Rosinach, À. Bravo, J. Deu-Pons, A. Bauer-Mehren, M. Baron, F. Sanz, and L. I. Furlong (2015) DisGeNET: A discovery platform for the dynamical exploration of human diseases and their genes. Database. 2015: bav028.

Ogata, H., S. Goto, K. Sato, W. Fujibuchi, H. Bono, and M. Kanehisa (1999) KEGG: Kyoto Encyclopedia of Genes and Genomes. Nucleic Acids Res. 27: 29–34.

Hewett, M., D. E. Oliver, D. L. Rubin, K. L. Easton, J. M. Stuart, R. B. Altman, and T. E. Klein (2002) PharmGKB: the pharmacogenetics knowledge base. Nucleic Acids Res. 30: 163-165.

Tate, J. G., S. Bamford, H. C. Jubb, Z. Sondka, D. M. Beare, N. Bindal, H. Boutselakis, C. G. Cole, C. Creatore, E. Dawson, P. Fish, B. Harsha, C. Hathaway, S. C. Jupe, C. Y. Kok, K. Noble, L. Ponting, C. C. Ramshaw, C. E. Rye, H. E. Speedy, R. Stefancsik, S. L. Thompson, S. Wang, S. Ward, P. J. Campbell, and S. A. Forbes (2019) COSMIC: the Catalogue Of Somatic Mutations In Cancer. Nucleic Acids Res. 47: D941-D947.

Lappalainen, I., J. Lopez, L. Skipper, T. Hefferon, J. D. Spalding, J. Garner, C. Chen, M. Maguire, M. Corbett, G. Zhou, J. Paschall, V. Ananiev, P. Flicek, and D. M. Church (2013) DbVar and DGVa: public archives for genomic structural variation. Nucleic Acids Res. 41: D936-D941.

Mailman, M. D., M. Feolo, Y. Jin, M. Kimura, K. Tryka, R. Bagoutdinov, L. Hao, A. Kiang, J. Paschall, L. Phan, N. Popova, S. Pretel, L. Ziyabari, M. Lee, Y. Shao, Z. Y. Wang, K. Sirotkin, M. Ward, M. Kholodov, K. Zbicz, J. Beck, M. Kimelman, S. Shevelev, D. Preuss, E. Yaschenko, A. Graeff, J. Ostell, and S. T. Sherry (2007) The NCBI dbGaP database of genotypes and phenotypes. Nat. Genet. 39: 1181-1186.

Smigielski, E. M., K. Sirotkin, M. Ward, and S. T. Sherry (2000) dbSNP: a database of single nucleotide polymorphisms. Nucleic Acids Res. 28: 352–355.

Liu, Z., M. Su, L. Han, J. Liu, Q. Yang, Y. Li, and R. Wang (2017) Forging the basis for developing protein-ligand interaction scoring functions. Acc. Chem. Res. 50: 302-309.

Su, M., Q. Yang, Y. Du, G. Feng, Z. Liu, Y. Li, and R. Wang (2019) Comparative assessment of scoring functions: The CASF-2016 update. J. Chem. Inf. Model. 59: 895-913.

Mysinger, M. M., M. Carchia, J. J. Irwin, and B. K. Shoichet (2012) Directory of useful decoys, enhanced (DUD-E): better ligands and decoys for better benchmarking. J. Med. Chem. 55: 6582-6594.

Carlson, H. A., R. D. Smith, K. L. Damm-Ganamet, J. A. Stuckey, A. Ahmed, M. A. Convery, D. O. Somers, M. Kranz, P. A. Elkins, G. Cui, C. E. Peishoff, M. H. Lambert, and J. B. Dunbar Jr. (2016) CSAR 2014: A benchmark exercise using unpublished data from pharma. J. Chem. Inf. Model. 56: 1063-1077.

Kim, S., J. Chen, T. Cheng, A. Gindulyte, J. He, S. He, Q. Li, B. A. Shoemaker, P. A. Thiessen, B. Yu, L. Zaslavsky, J. Zhang, and E. E. Bolton (2019) PubChem 2019 update: improved access to chemical data. Nucleic Acids Res. 47: D1102-D1109.

Mendez, D., A. Gaulton, A. P. Bento, J. Chambers, M. De Veij, E. Felix, M. P. Magarinos, J. F. Mosquera, P. Mutowo, M. Nowotka, M. Gordillo-Maranon, F. Hunter, L. Junco, G. Mugumbate, M. Rodriguez-Lopez, F. Atkinson, N. Bosc, C. J. Radoux, A. Segura-Cabrera, A. Hersey, and A. R. Leach (2019) ChEMBL: towards direct deposition of bioassay data. Nucleic Acids Res. 47: D930-D940.

Gilson, M. K., T. Liu, M. Baitaluk, G. Nicola, L. Hwang, and J. Chong (2016) BindingDB in 2015: A public database for medicinal chemistry, computational chemistry and systems pharmacology. Nucleic Acids Res. 44: D1045–1053.

Wishart, D. S., Y. D. Feunang, A. C. Guo, E. J. Lo, A. Marcu, J. R. Grant, T. Sajed, D. Johnson, C. Li, Z. Sayeeda, N. Assempour, I. Iynkkaran, Y. Liu, A. Maciejewski, N. Gale, A. Wilson, L. Chin, R. Cummings, D. Le, A. Pon, C. Knox, and M. Wilson (2018) DrugBank 5.0: a major update to the DrugBank database for 2018. Nucleic Acids Res. 46: D1074-D1082.

Kanehisa, M., M. Furumichi, M. Tanabe, Y. Sato, and K. Morishima (2017) KEGG: new perspectives on genomes, pathways, diseases and drugs. Nucleic Acids Res. 45: D353–D361.

Alexander, S. P. H., H. E. Benson, E. Faccenda, A. J. Pawson, J. L. Sharman, J. C. McGrath, W. A. Catterall, M. Spedding, J. A. Peters, A. J. Harmar, and CGTP Collaborators (2013) The concise guide to PHARMACOLOGY 2013/14: overview. Br. J. Pharmacol. 170: 1449-1458.

Hecker, N., J. Ahmed, J. von Eichborn, M. Dunkel, K. Macha, A. Eckert, M. K. Gilson, P. E. Bourne, and R. Preissner (2012) SuperTarget goes quantitative: update on drug-target interactions. Nucleic Acids Res. 40: D1113-D1117.

Gunther, S., M. Kuhn, M. Dunkel, M. Campillos, C. Senger, E. Petsalaki, J. Ahmed, E. G. Urdiales, A. Gewiess, L. J. Jensen, R. Schneider, R. Skoblo, R. B. Russell, P. E. Bourne, P. Bork, and R. Preissner (2008) SuperTarget and Matador: resources for exploring drug-target relationships. Nucleic Acids Res. 36: D919-D922.

Kuhn, M., C. von Mering, M. Campillos, L. J. Jensen, and P. Bork (2008) STITCH: interaction networks of chemicals and proteins. Nucleic Acids Res. 36: D684–D688.

Yang, H., C. Lou, L. Sun, J. Li, Y. Cai, Z. Wang, W. Li, G. Liu, and Y. Tang (2019) admetSAR 2.0: web-service for prediction and optimization of chemical ADMET properties. Bioinformatics. 35: 1067–1069.

Tomasulo, P. (2002) ChemIDplus-super source for chemical and drug information. Med. Ref. Serv Q. 21: 53–59.

Richard, A. M., R. S. Judson, K. A. Houck, C. M. Grulke, P. Volarath, I. Thillainadarajah, C. Yang, J. Rathman, M. T. Martin, J. F. Wambaugh, T. B. Knudsen, J. Kancherla, K. Mansouri, G. Patlewicz, A. J. Williams, S. B. Little, K. M. Crofton, and R. S. Thomas (2016) ToxCast chemical landscape: Paving the road to 21st century toxicology. Chem. Res. Toxicol. 29: 1225-1251.

Tox21 Challenge. https://tripod.nih.gov/tox21/challenge/ .

Watford, S., L. Ly Pham, J. Wignall, R. Shin, M. T. Martin, and K. P. Friedman (2019) ToxRefDB version 2.0: Improved utility for predictive and retrospective toxicology analyses. Reprod. Toxicol. 89: 145-158.

Sterling, T. and J. J. Irwin (2015) ZINC 15 — ligand discovery for everyone. J. Chem. Inf. Model. 55: 2324–2337

Blum, L. C. and J. L. Reymond (2009) 970 million druglike small molecules for virtual screening in the chemical universe database GDB-13. J. Am. Chem. Soc.131: 8732-8733.

Ruddigkeit, L., R. van Deursen, L. C. Blum, and J. L. Reymond (2012) Enumeration of 166 billion organic small molecules in the chemical universe database GDB-17. J. Chem. Inf. Model. 52: 2864–2875.

Ramakrishnan, R., P. O. Dral, M. Rupp, and O. A. von Lilienfeld (2014) Quantum chemistry structures and properties of 134 kilo molecules. Sci. Data. 1: 140022.

Visini, R., M. Awale, and J. L. Reymond (2017) Fragment database FDB-17. J. Chem. Inf. Model. 57: 700-709.

Sun, J., N. Jeliazkova, V. Chupakin, J. F. Golib-Dzib, O. Engkvist, L. Carlsson, J. Wegner, H. Ceulemans, I. Georgiev, V. Jeliazkov, N. Kochev, T. J. Ashby, and H. Chen (2017) ExCAPE-DB: an integrated large scale dataset facilitating Big Data analysis in chemogenomics. J. Cheminform. 9: 17.

Messenger, A. G. and J. Rundegren (2004) Minoxidil: Mechanisms of action on hair growth. Br. J. Dermatol. 150: 186–194.

Steinbach, G., P. M. Lynch, R. K. Phillips, M. H. Wallace, E. Hawk, G. B. Gordon, N. Wakabayashi, B. Saunders, Y. Shen, T. Fujimura, L. K. Su, B. Levin, L. Godio, S. Patterson, M. A. Rodriguez-Bigas, S. L. Jester, K. L. King, M. Schumacher, J. Abbruzzese, R. N. DuBois, W. N. Hittelman, S. Zimmerman, J. W. Sherman, and G. Kelloff (2000) The effect of celecoxib, a cyclooxygenase-2 inhibitor, in familial adenomatous polyposis. N. Engl. J. Med. 342: 1946-1952.

Von Eichborn, J., M. S. Murgueitio, M. Dunkel, S. Koerner, P. E. Bourne, and R. Preissner (2011) PROMISCUOUS: A database for network-based drug-repositioning. Nucleic Acids Res. 39: D1060–D1066.

Luo, H., P. Zhang, X. H. Cao, D. Du, H. Ye, H. Huang, C. Li, S. Qin, C. Wan, L. Shi, L. He, and L. Yang (2016) DPDR-CPI, a server that predicts drug positioning and drug repositioning via chemical-protein interactome. Sci. Rep. 6: 35996.

Brown, A. S. and C. J. Patel (2017) A standard database for drug repositioning. Sci. Data. 4: 170029.

Shameer, K., B. S. Glicksberg, R. Hodos, K. W. Johnson, M. A. Badgeley, B. Readhead, M. S. Tomlinson, T. O'Connor, R. Miotto, B. A. Kidd, R. Chen, A. Ma'ayan, and J. T. Dudley (2018) Systematic analyses of drugs and disease indications in RepurposeDB reveal pharmacological, biological and epidemiological factors influencing drug repositioning. Brief Bioinform. 19: 656–678.

Cotto, K. C., A. H. Wagner, Y. Y. Feng, S. Kiwala, A. C. Coffman, G. Spies, A. Wollam, N. C. Spies, O. L. Griffith, and M. Griffith (2018) DGIdb 3.0: A redesign and expansion of the drug-gene interaction database. Nucleic Acids Res. 46: D1068–D1073.

Kohler, S., L. Carmody, N. Vasilevsky, J. O. B. Jacobsen, D. Danis, J. P. Gourdine, M. Gargano, N. L. Harris, N. Matentzoglu, J. A. McMurry, D. Osumi-Sutherland, V. Cipriani, J. P. Balhoff, T. Conlin, H. Blau, G. Baynam, R. Palmer, D. Gratian, H. Dawkins, M. Segal, A. C. Jansen, A. Muaz, W. H. Chang, J. Bergerson, S. J. F. Laulederkind, Z. Yuksel, S. Beltran, A. F. Freeman, P. I. Sergouniotis, D. Durkin, A. L. Storm, M. Hanauer, M. Brudno, S. M. Bello, M. Sincan, K. Rageth, M. T. Wheeler, R. Oegema, H. Lourghi, M. G. Della Rocca, R. Thompson, F. Castellanos, J. Priest, C. Cunningham-Rundles, A. Hegde, R. C. Lovering, C. Hajek, A. Olry, L. Notarangelo, M. Similuk, X. A. Zhang, D. Gomez-Andres, H. Lochmuller, H. Dollfus, S. Rosenzweig, S. Marwaha, A. Rath, K. Sullivan, C. Smith, J. D. Milner, D. Leroux, C. F. Boerkoel, A. Klion, M. C. Carter, T. Groza, D. Smedley, M. A. Haendel, C. Mungall, and P. N. Robinson (2019) Expansion of the Human Phenotype Ontology (HPO) knowledge base and resources. Nucleic Acids Res. 47: D1018–D1027.

Download references

Acknowledgments

This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (NRF-2020R1A2C2004628), and was supported by the Bio-Synergy Research Project (NRF-2017M3A9C 4092978) of the Ministry of Science, ICT.

Author information

Authors and affiliations.

School of Electrical Engineering and Computer Science, Gwangju Institute of Science and Technology (GIST), Gwangju, 61005, Korea

Hyunho Kim, Eunyoung Kim, Ingoo Lee, Bongsung Bae, Minsu Park & Hojung Nam

You can also search for this author in PubMed   Google Scholar

Corresponding author

Correspondence to Hojung Nam .

Additional information

Publisher’s note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

The authors declare no conflict of interest.

Neither ethical approval nor informed consent was required for this study.

Rights and permissions

Reprints and permissions

About this article

Kim, H., Kim, E., Lee, I. et al. Artificial Intelligence in Drug Discovery: A Comprehensive Review of Data-driven and Machine Learning Approaches. Biotechnol Bioproc E 25 , 895–930 (2020). https://doi.org/10.1007/s12257-020-0049-y

Download citation

Received : 13 February 2020

Revised : 27 May 2020

Accepted : 03 June 2020

Published : 07 January 2021

Issue Date : December 2020

DOI : https://doi.org/10.1007/s12257-020-0049-y

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• drug discovery
• artificial intelligence
• data-driven
• machine learning

Advertisement

• Find a journal
• Publish with us
• Track your research

MIT Technology Review

• Newsletters

Google DeepMind’s new AlphaFold can model a much larger slice of biological life

AlphaFold 3 can predict how DNA, RNA, and other molecules interact, further cementing its leading role in drug discovery and research. Who will benefit?

• James O'Donnell archive page

Google DeepMind has released an improved version of its biology prediction tool, AlphaFold, that can predict the structures not only of proteins but of nearly all the elements of biological life.

It’s a development that could help accelerate drug discovery and other scientific research. The tool is currently being used to experiment with identifying everything from resilient crops to new vaccines. 

While the previous model, released in 2020, amazed the research community with its ability to predict proteins structures, researchers have been clamoring for the tool to handle more than just proteins. 

Now, DeepMind says, AlphaFold 3 can predict the structures of DNA, RNA, and molecules like ligands, which are essential to drug discovery. DeepMind says the tool provides a more nuanced and dynamic portrait of molecule interactions than anything previously available. 

“Biology is a dynamic system,” DeepMind CEO Demis Hassabis told reporters on a call. “Properties of biology emerge through the interactions between different molecules in the cell, and you can think about AlphaFold 3 as our first big sort of step toward [modeling] that.”

AlphaFold 2 helped us better map the human heart , model antimicrobial resistance , and identify the eggs of extinct birds , but we don’t yet know what advances AlphaFold 3 will bring. 

Mohammed AlQuraishi, an assistant professor of systems biology at Columbia University who is unaffiliated with DeepMind, thinks the new version of the model will be even better for drug discovery. “The AlphaFold 2 system only knew about amino acids, so it was of very limited utility for biopharma,” he says. “But now, the system can in principle predict where a drug binds a protein.”

Isomorphic Labs, a drug discovery spinoff of DeepMind, is already using the model for exactly that purpose, collaborating with pharmaceutical companies to try to develop new treatments for diseases, according to DeepMind. 

AlQuraishi says the release marks a big leap forward. But there are caveats.

“It makes the system much more general, and in particular for drug discovery purposes (in early-stage research), it’s far more useful now than AlphaFold 2,” he says. But as with most models, the impact of AlphaFold will depend on how accurate its predictions are. For some uses, AlphaFold 3 has double the success rate of similar leading models like RoseTTAFold. But for others, like protein-RNA interactions, AlQuraishi says it’s still very inaccurate. 

DeepMind says that depending on the interaction being modeled, accuracy can range from 40% to over 80%, and the model will let researchers know how confident it is in its prediction. With less accurate predictions, researchers have to use AlphaFold merely as a starting point before pursuing other methods. Regardless of these ranges in accuracy, if researchers are trying to take the first steps toward answering a question like which enzymes have the potential to break down the plastic in water bottles, it’s vastly more efficient to use a tool like AlphaFold than experimental techniques such as x-ray crystallography. 

A revamped model  

AlphaFold 3’s larger library of molecules and higher level of complexity required improvements to the underlying model architecture. So DeepMind turned to diffusion techniques, which AI researchers have been steadily improving in recent years and now power image and video generators like OpenAI’s DALL-E 2 and Sora. It works by training a model to start with a noisy image and then reduce that noise bit by bit until an accurate prediction emerges. That method allows AlphaFold 3 to handle a much larger set of inputs.

That marked “a big evolution from the previous model,” says John Jumper, director at Google DeepMind. “It really simplified the whole process of getting all these different atoms to work together.”

It also presented new risks. As the AlphaFold 3 paper details, the use of diffusion techniques made it possible for the model to hallucinate, or generate structures that look plausible but in reality could not exist. Researchers reduced that risk by adding more training data to the areas most prone to hallucination, though that doesn’t eliminate the problem completely. 

Restricted access

Part of AlphaFold 3’s impact will depend on how DeepMind divvies up access to the model. For AlphaFold 2, the company released the open-source code , allowing researchers to look under the hood to gain a better understanding of how it worked. It was also available for all purposes, including commercial use by drugmakers. For AlphaFold 3, Hassabis said, there are no current plans to release the full code. The company is instead releasing a public interface for the model called the AlphaFold Server , which imposes limitations on which molecules can be experimented with and can only be used for noncommercial purposes. DeepMind says the interface will lower the technical barrier and broaden the use of the tool to biologists who are less knowledgeable about this technology.

Artificial intelligence

What’s next for generative video.

OpenAI's Sora has raised the bar for AI moviemaking. Here are four things to bear in mind as we wrap our heads around what's coming.

• Will Douglas Heaven archive page

Is robotics about to have its own ChatGPT moment?

Researchers are using generative AI and other techniques to teach robots new skills—including tasks they could perform in homes.

• Melissa Heikkilä archive page

Sam Altman says helpful agents are poised to become AI’s killer function

Open AI’s CEO says we won’t need new hardware or lots more training data to get there.

An AI startup made a hyperrealistic deepfake of me that’s so good it’s scary

Synthesia's new technology is impressive but raises big questions about a world where we increasingly can’t tell what’s real.

Stay connected

Get the latest updates from mit technology review.

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at [email protected] with a list of newsletters you’d like to receive.

• Open access
• Published: 09 May 2024

Exploiting urine-derived induced pluripotent stem cells for advancing precision medicine in cell therapy, disease modeling, and drug testing

• Xiya Yin 1 , 2 ,
• Qingfeng Li 1 ,
• Yan Shu 3 ,
• Hongbing Wang 3 ,
• Biju Thomas 4 ,
• Joshua T. Maxwell 5 &
• Yuanyuan Zhang   ORCID: orcid.org/0000-0002-5708-9718 5  

Journal of Biomedical Science volume  31 , Article number:  47 ( 2024 ) Cite this article

187 Accesses

Metrics details

The field of regenerative medicine has witnessed remarkable advancements with the emergence of induced pluripotent stem cells (iPSCs) derived from a variety of sources. Among these, urine-derived induced pluripotent stem cells (u-iPSCs) have garnered substantial attention due to their non-invasive and patient-friendly acquisition method. This review manuscript delves into the potential and application of u-iPSCs in advancing precision medicine, particularly in the realms of drug testing, disease modeling, and cell therapy. U-iPSCs are generated through the reprogramming of somatic cells found in urine samples, offering a unique and renewable source of patient-specific pluripotent cells. Their utility in drug testing has revolutionized the pharmaceutical industry by providing personalized platforms for drug screening, toxicity assessment, and efficacy evaluation. The availability of u-iPSCs with diverse genetic backgrounds facilitates the development of tailored therapeutic approaches, minimizing adverse effects and optimizing treatment outcomes. Furthermore, u-iPSCs have demonstrated remarkable efficacy in disease modeling, allowing researchers to recapitulate patient-specific pathologies in vitro. This not only enhances our understanding of disease mechanisms but also serves as a valuable tool for drug discovery and development. In addition, u-iPSC-based disease models offer a platform for studying rare and genetically complex diseases, often underserved by traditional research methods. The versatility of u-iPSCs extends to cell therapy applications, where they hold immense promise for regenerative medicine. Their potential to differentiate into various cell types, including neurons, cardiomyocytes, and hepatocytes, enables the development of patient-specific cell replacement therapies. This personalized approach can revolutionize the treatment of degenerative diseases, organ failure, and tissue damage by minimizing immune rejection and optimizing therapeutic outcomes. However, several challenges and considerations, such as standardization of reprogramming protocols, genomic stability, and scalability, must be addressed to fully exploit u-iPSCs’ potential in precision medicine. In conclusion, this review underscores the transformative impact of u-iPSCs on advancing precision medicine and highlights the future prospects and challenges in harnessing this innovative technology for improved healthcare outcomes.

Introduction

Initially, stem cell research was primarily centered on harnessing the potential of ESCs, which were derived from human embryos. Despite the tremendous promise of ESCs, their use became mired in ethical and political controversy due to the necessity of embryo destruction for their procurement [ 1 , 2 ]. However, in 2006, a momentous breakthrough emerged from Kyoto University in Japan, led by Shinya Yamanaka and his team. They achieved a transformative feat by successfully reprogramming adult mouse fibroblast cells into a pluripotent state by introducing a set of four specific transcription factors: Oct4, Sox2, Klf4, and c-Myc. These reprogrammed cells earned the name ‘induced pluripotent stem cells’ (iPSCs) [ 3 ]. Building on this achievement, Yamanaka’s team extended their reprogramming method to human cells in 2007, using the same quartet of transcription factors to create human iPSCs [ 4 ]. Induced pluripotent stem cells (iPSCs) are pluripotent stem cells generated from patients’ own somatic cells by reprogramming them to an embryonic stem cell-like state, making it possible to create a patient-specific disease model customized for their genetic information without raising similar ethical considerations of embryonic stem cells (ESCs) [ 5 ]. iPSCs have the potential to differentiate into three germ layers and generate a broad spectrum of cell types in the body, suitable as an appreciable tool for regenerative medicine. Meanwhile, the patient-specific origin of iPSCs leads to the eliminated risk of immune rejection and enhanced effectiveness while used in autologous cell therapy. iPSCs can be generated from diverse readily procurable cell sources like dermal fibroblasts and peripheral blood mononuclear cells (PBMCs) from an extensive variety of populations, expanding the scope of research models [ 6 , 7 ]. iPSCs show unique values in investigating rare and genetic diseases when relevant animal models are scarce, as the introduction of specific genetic mutations assists in studying the progression of diseases related to individuals [ 8 ]. In brief, iPSCs demonstrate massive benefits in regenerative medicine, drug discovery, and disease modeling.

Despite the merits, there are several challenges in the research of iPSCs. The efficient induction of iPSCs requires rigorously regularized protocols, which has not reached a common sense among diverse laboratories. The incomplete differentiation of iPSCs may cause the formation of teratomas, provoking uncertainties regarding the safety issue in the utilization of iPSCs. Additionally, the generation of iPSCs touches on ethical and regulatory apprehensions concerning the invasive acquisition procedure of cell sources from patients [ 8 ]. From the aspect of clinical transition, the determination of an easily obtainable cell source for large-scale production, and ways to minimize the cost for iPSC generation and differentiation remain unsettled troubles.

A proper cell source is vital for the generation and application of iPSC. The cell source should be easily acquirable from consistent donors, which is more reproducible and reliable with fewer risks of variability. Regarding cell features, the cell source with high reprogram ability and proliferative capacity is more efficient for the generation of iPSC. To ensure its safety in patients, the cell source should possess genetic and epigenetic stability, with no tendency of tumorigenicity. In addition, the tissue origin is pivotal due to the related obtaining process, ethical concerns, and compatibility with research goals.

Urine-derived stem cells (USCs) are a type of adult stem cell derived from body fluids that can be isolated from urine samples, a waste product that is routinely discarded [ 9 ]. Unlike skin or blood cells, the non-invasive collection of USCs eliminates the need for invasive surgical procedures, reducing patient discomfort and potential risks associated with tissue harvesting. USCs have demonstrated excellent proliferative capacity, self-renewal ability, immunomodulatory properties [ 10 , 11 ], and the ability to differentiate into various cell types, including but not limited to neurons, bone cells, muscle cells, and cartilage cells [ 12 , 13 ]. These unique characteristics make USCs particularly advantageous as a cell source for iPSCs.

Urine-derived iPSCs (u-iPSCs) are a subtype of iPSCs generated from cells in urine samples [ 14 ], mainly from USCs [ 15 , 16 ]. As a cell source of patient-specific pluripotent cells, u-iPSCs can be obtained at low cost with a non-invasive and patient-friendly method and offer several advantages over other sources of iPSCs. USCs can be reprogrammed into iPSCs and then differentiated into various cell types. USCs reprogram into iPSCs more efficiently and rapidly than other somatic cells. They achieve of 80% transduction rate compared to 50% in mesenchymal cell lines. USC-derived iPSCs show morphological changes indicative of reprogramming within 3 days, form distinct colonies expressing pluripotency markers by day 7, and reach maturity by day 10–14, whereas mesenchymal cell-derived colonies require 28 days [ 17 ]. The shorter induction time and higher reprogramming efficiency owes to epithelial origin of USCs, which means the elimination of mesenchymal-to-epithelial transition (MET) process [ 12 , 16 , 17 ]. U-iPSCs have been adopted in the research of precision medicine, especially the establishment of patient-specific disease modelling, including neuromuscular, neurodegenerative, cardiovascular, hematopoietic, and pediatric diseases.

Advancing technology for the generation of u-iPSCs

Over the subsequent years, the realm of iPSC research witnessed rapid expansion and diversification (Fig.  1 ). Researchers devised a variety of techniques for generating iPSCs, incorporating alternative transcription factors and non-viral delivery approaches. These innovations significantly improved the efficiency and safety of iPSCs production [ 18 ].

Advancing techniques in utilizing u-iPSCs. 3D printing techniques, organoids and gene editing techniques are three emerging technologies in the study of u-iPSCs. iPSCs can be used in 3D bioprinting to create patient-specific organoids for transplantation, which could address the shortage of organ donors and reduce the risk of transplant rejection. Moreover, gene editing techniques like CRISPR/Cas9 can be used to correct genetic mutations in iPSCs before they are used in therapeutic applications. (Created with BioRender.com)

Gene-editing techniques

Despite the reprogramming methods mentioned above, new technologies have been applied in the research of u-iPSC, such as using genome editing to directly correct genetic mutations and develop novel gene therapy methods. Gene editing technologies, such as CRISPR/Cas9 system [ 3 , 4 , 19 ], zinc-finger nuclease (ZFN) [ 18 , 20 , 21 ], and transcription activator-like effector nucleases (TALENs) [ 22 ], allow the direct insertion, deletion, or replacement of distinct DNA sequences, creating specific modification in patients’ genome at targeted locations. Recent studies have focused on generating gene-edited urine-derived induced pluripotent stem cells (u-iPSCs) from patients with various diseases, aiming to explore novel gene therapy approaches. Zou et al. successfully generated u-iPSCs from the urine of an achondroplasia (ACH) patient and corrected the Gly380Arg mutation using CRISPR-Cas9, thereby restoring the chondrogenic differentiation ability of ACH iPSCs [ 23 ]. Similarly, Zhou et al. generated u-iPSCs from spinal muscular atrophy (SMA) patients and converted the survival motor neuron 2 (SMN2) gene to a survival motor neuron 1 (SMN1)-like gene using CRISPR/Cpf1 and single-stranded oligodeoxynucleotides (ssODN). The resulting motor neurons (iMNs) from modified u-iPSCs exhibited rescued expression of SMN proteins [ 24 ]. In addition, Neumeyer et al. used the piggyBac DNA transposon system to integrate the human F8 gene into the genome of u-iPSCs derived from individuals with hemophilia A. Upon differentiation of the modified u-iPSCs into endothelial cells, they formed vascular networks and demonstrated the capacity to produce functional FVIII when implanted into the subcutaneous tissue of hemophilic mice [ 25 ]. Furthermore, Zeng et al. reprogrammed urinary stem cells (USCs) collected from Duchenne muscular dystrophy (DMD) patients with an exon 50 deletion into u-iPSCs [ 26 ]. Subsequently, they used TALEN-based nickases to integrate a functional mini-dystrophin gene into the rDNA locus of the u-iPSCs. Mini-dystrophin expression was detected both in the genetically modified u-iPSCs and in the cardiomyocytes differentiated from them.

It’s worth noting that due to the non-specific action of gene editing tools, they may inadvertently modify DNA regions that are like, but not the intended target gene. This can lead to the generation of unknown genetic variations. While researchers can use DNA sequencing to determine whether changes have occurred in off-target regions, there remains a certain level of risk to patient safety in clinical applications. To better protect patients, the precision and specificity of gene editing tools have been continuously improved and refined [ 4 , 27 ]. The use of gene editing technology is also subject to strict legal regulations and ethical considerations.

Dissimilar to traditional two-dimensional (2D) in vitro cell culture which is incapable of mimicking the natural environment in vivo, an organoid is sophisticated designed using various cells, possessing the three-dimensional (3D) structure of a tissue or organ, reflecting the complex cell-cell communications and tissue interactions, and resembling in vivo functions [ 28 ]. iPSC-based organoids can be generated with cells obtained from specific patients, which provides a disease-targeted model for researchers to conduct a multitude of experiments on the underlying mechanisms of specific disease.

U-iPSC [ 29 ], as a non-invasive, readily available, consistent cell source with high proliferation capacity, reprogramming efficiency, low risks, and no ethical consequences, have been applied in the field of organoid research. Mulder et al. induced u-iPSCs from infant and pediatric urine with episomal vectors and generated human kidney organoids after rigorously characterization of their pluripotency and karyotyping [ 30 ]. Kim et al. investigated the effect of Matrigel and Y-27,632 on promoting self-renewal and differentiation capacity of USCs and successfully generated kidney organoid and hematopoietic progenitor cells from u-iPSCs [ 31 ]. To investigate the pathophysiological mechanisms of glomerular diseases, a u-iPSC based kidney organoid was developed by Nguyen et al. with artificially induced injuries using puromycin aminonucleosides (PAN) [ 32 ]. An interconnected network related to inflammation and cell death was confirmed, revealing the potential of u-iPSC based kidney organoid in regenerative medicine for kidney diseases.

Despite kidney organoids, u-iPSC have been utilized in the development of retinal organoids [ 33 ], microvascular grafts [ 25 ], cerebral organoids [ 34 ] and tooth-like structures [ 35 ]. Li et al. formed 3D retinal organoids with properly layered neural retina containing all retinal cell types by differentiating u-iPSCs into retinal fates. Notably, u-iPSCs produced highly mature photoreceptors, including red/green cone-rich photoreceptors, without the supplementation of retinoic acid [ 33 ].

Neumeyer et al. genetically modified u-iPSCs with full-length F8 and differentiated them into endothelial cells (ECs). These cells produced high levels of FVIII and self-assembled into vascular networks upon subcutaneous implantation into hemophilic mice, effectively correcting the clotting deficiency and offering a potential autologous ex vivo gene-therapy strategy for HA treatment [ 25 ]. Teles et al. generated three-dimensional human cerebral organoids with neurons and astrocytes differentiated from u-iPSCs derived from Down syndrome (DS) patients [ 34 ], demonstrating the developmental dynamics of the early-stage forebrain. In the study by Cai et al., u-iPSCs were differentiated into epithelial sheets and combined with mouse dental mesenchymes, resulting in tooth-like structures within 3 weeks with a success rate of up to 30% across 8 iPSC lines, comparable to hESCs. These structures contained enamel-secreting ameloblasts with physical properties resembling human teeth [ 35 ].

3D bioprinting technology

The combination of stem cells with other emerging technologies, such as 3D bioprinting and nanotechnology, has been adopted to create novel regenerative medicine strategies. iPSCs, derived from a patient’s own dermal fibroblasts or peripheral blood mononuclear cells, offer a sustainable source of cells for 3D printing. Scientists have achieved success in utilizing bio-inks containing human iPSCs to 3D print a wide array of tissues and organs, including but not limited to cartilage [ 36 ], skin [ 37 ], heart, liver [ 38 ], and neural tissues [ 39 ]. These tailored biological constructs not only cater to individual patient needs but also account for their unique genetic variations, thereby markedly reducing the risk of rejection. They hold substantial promise for playing a more prominent role in tissue repair and regeneration. In the realm of 3D printing technology, it can create intricate structures of tissues and organ models, faithfully replicating the microenvironments found in actual human diseases. Consequently, this technology is instrumental in disease modeling. In the domain of cardiovascular diseases, iPSC-derived cardiac cells have proven to be effective in emulating conditions such as dilated cardiomyopathy and myocardial infarction [ 40 ]. Within the context of neurodegenerative diseases, the utilization of 3D bioprinting with iPSCs has given rise to disease models for conditions such as Alzheimer’s [ 41 ], Parkinson’s [ 42 ], and amyotrophic lateral sclerosis (ALS).

These models serve to explore the interactions among various types of nerve cells, decipher the pathophysiological characteristics of diseases, and delve into the mechanisms underlying disease onset. For oncological diseases, 3D bioprinting with cell lines derived from iPSCs can construct structures resembling tumors, such as spheroids or organoids, faithfully simulating the tumor microenvironment. These models serve as a platform for investigating the different stages of cancer progression following transplantation into animal models [ 43 ]. Moreover, artificial skin models created through 3D bioprinting, using iPSCs as a foundation, maintain intricate cellular pathways, interactions between cells, and the interplay between cells and their microenvironment [ 44 ]. This attribute confers substantial research value, especially in the realms of drug toxicity testing and the evaluation of cosmetic products.

Using 3D-printing technology, USCs have been combined with various biomaterials as a construction and applied to the research of bone tissue regeneration and repair. While 3D-printing technology allows for personalized bone substitutes, it lacks the ability to regulate the topological morphology of the scaffold surface, which is crucial for stem cell behavior. The fabricated poly(e-caprolactone) (PCL) scaffold with nanoridge patterns constructed by Xing et al. enhanced protein adsorption and mineralization compared to bare PCL scaffolds. Loaded with USCs, these scaffolds showed increased proliferation, cell length, and osteogenic gene expression, indicating improved bone regeneration capability [ 45 ]. Zhang et al. built a 3D-printed polylactic acid and hydroxyapatite (PLA/HA) composite scaffold loaded with USCs in treating skull defects in a rat model. Evaluation at 4, 8, and 12 weeks revealed that the PLA/HA scaffold with USCs significantly promoted new bone regeneration, with nearly complete coverage of the defect area observed at 12 weeks. These results underscore the potential of 3D-printed scaffolds with USCs in bone tissue engineering [ 46 ]. For now, studies considering u-iPSCs as a cell source for 3D printing remain scarce. Shao et al. successfully generated u-iPSCs and differentiated them into neural stem cells (NSCs). The 3D printed scaffold loaded with these NSCs showed preferable efficacy in repairing spinal cord injury after transplanted into mouse models, indicating the potential of u-iPSC in tissue regeneration and repair [ 47 ]. We look forward to more related research to further confirm the application value of u-iPSCs.

Applications of u-iPSCs in precision medicine

U-iPSCs have the potential to revolutionize precision medicine by enabling personalized approaches to cell therapy, drug testing, and disease modeling ( Fig.  2 ).

Disease modelling using u-iPSCs. USCs are harvested from urine of patients with specific mutations, and reprogrammed into u-iPSCs, which reflects the pathological condition under laboratory settings. Gene editing tools, such as CRISPR/Cas9, can be used to correct the genetic mutation in the patient’s u-iPSCs. The modified u-iPSCs are subsequently used in disease modelling, drug discovery, cell therapy and biomarker identification. (Created with BioRender.com)

• Cell therapy

In the field of regenerative medicine, stem cells are widely used for tissue repair, regeneration, and cell therapy for various diseases due to their ability to self-renew and differentiate into various specific cell types [ 48 ]. Our team have previously published a series of studies where the efficacy and mechanism of USC-based cell therapy in various diseases such as Type 2 diabetic erectile dysfunction and complications [ 49 , 50 , 51 , 52 ], bladder diseases [ 53 , 54 ], acute and chronic kidney injuries [ 55 , 56 ], male infertility [ 2 , 57 ], and inflammatory bowel diseases [ 11 ]. Nonetheless, adult stem cells have limited differentiation ability, and embryonic stem cells, despite having high differentiation potential, face ethical limitations in their acquisition process, hindering their clinical applications [ 58 ].

iPSCs, on the other hand, can be generated from various sources such as the patient’s own skin, blood, urine, etc. They possess the ability to differentiate into various cell types representing all three germ layers. Additionally, iPSCs exhibit low immunogenicity and do not involve the ethical concerns associated with embryonic stem cells, making them a novel tool for stem cell therapy research [ 59 ]. U-iPSCs have been generated from patients’ urine and differentiated into various cell types, which enables the development of patient-specific cell replacement therapies for degenerative diseases, organ failure, and tissue damage. Apart from the fundamental osteogenic, chondrogenic, and adipogenic capacity, U-iPSCs can be differentiated into alveolar type II epithelial cells [ 60 ], cardiomyocytes [ 61 , 62 , 63 ], fibroblasts and skeletal muscle myocytes [ 64 ], epithelial cells [ 35 , 65 ], neurons and astrocytes [ 15 , 24 , 34 , 65 , 66 ], hepatocyte-like cells [ 67 , 68 ], lens progenitor cells [ 69 ], retinal cell [ 33 ], and kidney precursor cells [ 70 ]. There have been studies on the utilization of u-iPSCs and their differentiated cells in the research of cell therapies of renal and neurological diseases.

Kidney disease encompasses two main types: chronic and acute. Acute kidney disease is characterized by a rapid decline in kidney function over a short period, often caused by severe infections, ischemia, drug toxicity, and other factors. Patients may experience symptoms such as oliguria, nausea, and vomiting. Chronic kidney disease, on the other hand, develops over the long term due to conditions like prolonged high blood pressure, diabetes, etc. It is a progressive condition, and patients may exhibit symptoms like fatigue, decreased appetite, and edema [ 71 ]. For those in end-stage renal disease, kidney transplantation is an effective treatment; however, challenges such as donor shortage, immune rejection, surgical complications, etc., exist [ 72 ]. Therefore, stem cell replacement therapy has emerged as a promising new approach.

Diabetic nephropathy is a form of chronic kidney disease caused by diabetes, which may eventually lead to kidney failure. Gao et al. generated u-iPSCs from urine sample of patients with diabetic nephropathy and directed their differentiation into induced nephron progenitor cells (iNPCs), which were subsequently injected into cortex of the diabetic mice’s kidney. The findings suggested that these u-iPSC derived iNPCs presented significant efficacy with reduced inflammation and fibrosis, promoted kidney regeneration and improved renal function [ 73 ]. Concerning acute kidney injury (AKI), Jin et al. established u-iPSCs from AKI patients and directed the differentiation into kidney precursor cells (KPCs). After transplantation into an ischemia–reperfusion-induced AKI mice model, the renal function was significantly ameliorated, reflected by the improvement of reduced serum creatinine and BUN levels [ 70 ].

Stress urinary incontinence (SUI) is common in women and the elderly, referring to the leakage of urine caused by an increase in abdominal pressure during activities such as coughing, sneezing, or engaging in sports. The main reason patients cannot control urine on their own is the loss of tension or dysfunction of pelvic floor muscles and the urethral sphincter due to factors such as childbirth, age, and obesity [ 74 ]. Urinary incontinence may affect the normal work and social life of patients, and the costs associated with rehabilitation and nursing services also impose a certain economic burden. Stem cell therapy may promote the repair and regeneration of damaged tissues by directing differentiation, anti-inflammatory effects, and secretion of neuroprotective factors. This approach could help improve the function of the urethral sphincter, thereby enhancing the patient’s ability to control urine. Kibschull et al. established u-iPSCs from urine of female SUI patients and differentiated them into fibroblasts and myocytes. At the three-week time point after periurethral injection into rats, these differentiated cells were traceable and found active in the periurethral areas, showing their feasibility in urethral repair and regeneration [ 64 ].

Spinal cord injury (SCI) refers to the structural and functional damage to the spinal cord caused by trauma or disease, often resulting in sensory, motor, and autonomic nervous system impairments in the areas it innervates. Severe cases may lead to disability [ 75 ]. Currently, apart from symptomatic treatment and rehabilitation measures, there is no cure for spinal cord injuries. In recent years, researchers have been exploring the role of stem cell therapy in promoting the repair and regeneration of the spinal cord [ 76 ]. Liu et al. generated neural progenitor cells (NPCs) with human u-iPSCs and transplanted NPCs into the neural tissues adjacent to lesion site of SCI rat model. The accumulation of u-iPSCs derived NPCs were observed at the lesion cavity and some differentiated into neurons, astrocytes, or oligodendrocytes, confirming the potential of u-iPSCs in nerve repair and regeneration [ 66 ].

In addition to iPSCs derived from urine, iPSCs from other cell sources have also shown potential in cell therapy for various diseases, especially neurodegenerative disorders, as demonstrated in a series of animal experiments and clinical trials [ 77 , 78 , 79 ]. However, before their widespread clinical application, there are still some issues related to the inherent characteristics of iPSCs that need to be addressed.

• Drug testing

Drug screening

U-iPSCs can be used to generate patient-specific disease models, which can then be used to screen drugs for efficacy and toxicity in a personalized manner. This can help to identify the most effective and safest treatments for individual patients. There are primarily two methods for drug screening: target-based drug screening and phenotype-based drug screening. Target-based screening is predominantly employed when there is a comprehensive understanding of the disease mechanism, and specific key enzymes, proteins, or receptors have been pinpointed. In this approach, drugs are administered with precision to evaluate their impacts on these biological molecular targets. This screening method enhances our comprehension of the precise mechanisms of drugs and aids researchers in fine-tuning drug candidates [ 80 ]. Nevertheless, as the targets often originate from idealized laboratory research models, their applicability to the intricate human body environment may be limited.

In contrast to the traditional target-based approach, phenotype-based drug screening is principally appropriate for diseases with insufficiently understood mechanisms. By observing alterations in cellular phenotypes or functions after drug treatment, the objective is to identify a drug that accomplishes the desired effects for subsequent validation and refinement. Although there are certain challenges associated with investigating specific molecular mechanisms, this method is of great value for diseases where the underlying mechanisms remain incompletely elucidated [ 81 ]. However, drugs identified through phenotype-based screening from cellular or animal models align more closely with the underlying pathological and physiological nature of the disease. This not only enhances efficiency but also increases the feasibility of drug discovery [ 82 ]. Since the introduction of iPSCs, there have been significant advancements in phenotype-based screening. Researchers achieve this by reprogramming a patient’s somatic cells into iPSCs, utilizing gene editing techniques to correct disease-associated loci within the cellular genes, thereby creating isogenic control models. After guiding iPSCs to differentiate into disease-specific cell types, drug treatments are administered, and subsequent observations of phenotypic changes in disease and control models are made to identify effective therapeutic agents [ 83 ].

u-iPSCs can be utilized to create cell lines that replicate disease phenotypes specific to individual patients, ensuring a reliable source of cell types that were previously difficult to access and expand, including neurons and cardiomyocytes. They are well-suited for conducting high-throughput drug screening to evaluate the efficacy of a wide range of pharmaceuticals. For instance, there have been studies using neural precursor cells derived from iPSCs of patients with Fragile X syndrome (FXS) to identify effective compounds which increase the expression of deficient proteins, thus providing positive proofs for FXS drug development [ 84 , 85 , 86 ]. Niemietz et al. generated u-iPSCs from familial amyloid polyneuropathy (FAP) patients and directed in vitro differentiation into hepatocytes. The knockdown of FAP related mutation gene transthyretin (TTR) with therapeutic oligonucleotides in u-iPSC derived hepatocytes presents high efficiency, confirming u-iPSCs as a useful tool for novel compounds screening of FAP [ 12 ]. iPSCs contribute significantly to in-depth comprehension of drug mechanisms and the identification of relevant drug targets. Furthermore, they help reduce ethical concerns related to animal experimentation while improving the efficiency of research. In general, iPSCs offer numerous advantages for phenotype-based screening. Further research is needed for the application of u-iPSCs in this area.

Toxicity screening

In the drug development process, comprehensive and thorough testing of drug reactivity, activity, and toxicity is crucial to ensure the effectiveness and safety of drugs once they enter the market. In the United States, approximately 60% of hospitalized patients with acute kidney injury are related to drug-induced kidney toxicity [ 87 ]. The annual socioeconomic burden resulting from drug-induced kidney toxicity can be as high as 900 million dollars [ 88 ]. The mechanisms of drug-induced kidney toxicity are complex and wide-ranging, involving various target sites such as renal tubular epithelial cells, podocytes, renal interstitium, microvascular systems [ 89 ]. In the research of drug-induced kidney toxicity mechanisms, traditional in vitro 2D cell culture models cannot effectively reflect the interactions between cells and the extracellular matrix in the in vivo microenvironment. Additionally, as cells undergo passaging, their phenotype and function may change, affecting the effectiveness and accuracy of toxicity testing. Animal models are costly, time-consuming, and raise ethical concerns. Furthermore, the differences in disease-related protein and enzyme expression between animals and humans can impact the clinical utility of drugs.

In vitro 3D culture models, such as organoids and engineered kidney tissues, consist of a diverse array of renal cell types and feature three-dimensional spatial arrangements that closely mimic the real physiological environment. Therefore, they are better suited for drug toxicity testing [ 90 ]. Our team co-cultivated USCs with Kidney-specific ECM to construct USC organoids resembling renal tubules and kidney-like organoids. Upon examination, these USC organoids exhibited a compact 3D structure with minimal central necrosis and high cell viability. They expressed specific markers such as Aquaporin-1 (AQP1) for proximal tubules, Podocin and Synaptopodin for renal glomeruli, and the secretion of erythropoietin (EPO) by renal interstitial cells. The results of drug toxicity testing showed that USC organoids were responsive to nephrotoxic drugs such as aspirin, penicillin G, acetone, and cisplatin, resulting in cell necrosis [ 91 , 92 ]. Therefore, in vitro USC organoids constructed in this manner can simulate the phenotype and function of the kidney, making them suitable for studying the actual effects of drugs in a physiological environment. Kidney-like organs derived from iPSC sources contain a greater variety of cell types at different developmental stages. Further development of organoids based on u-iPSCs may result in models that are more suitable for toxicity testing [ 93 ].

Mitochondrial dysfunction plays a significant role in the mechanisms of drug toxicity. Drugs induce damage and functional impairment of mitochondria through various mechanisms, including inhibiting mitochondrial replication, affecting the electron transport chain responsible for ATP synthesis, altering mitochondrial permeability, and inhibiting the function of mitochondrial membrane transport proteins. In highly metabolic organs like the heart, kidney, skeletal muscle, and in the liver when drug concentrations are high, drug-induced mitochondrial damage not only leads to organ toxicity but can also result in symptoms such as increased glycolysis and lactic acid accumulation, leading to acidosis [ 94 ]. Therefore, the assessment of in vivo and in vitro mitochondrial toxicity is an essential component of drug safety evaluation in the drug development process. We developed 3D USC spheroids to assess the chronic cytotoxicity and mitochondrial toxicity of anti-retroviral drugs, including zalcitabine, tenofovir, and Raltegravir. The results showed that these drugs inhibited the expression of certain mitochondrial oxidative phosphorylation enzymes and reduced mitochondrial DNA content [ 95 ]. Furthermore, we seeded USCs onto silk fibers to construct three-dimensional tissue-engineered structures and treated them with anti-retroviral drugs. The results demonstrated that this model could more sensitively reflect the effects of drugs on mitochondria compared to 3D USC spheroids [ 96 ] (Fig.  3 ).

The process of drug discovery with patient-specific u-iPSCs. After the generation from patients’ urine, U-iPSCs are differentiated into specific cell types relevant to the disease. Subsequently, the high-throughput screening is conducted, where thousands of chemical compounds or drugs are rapidly tested to identify potential candidates that have a desired effect on the cells. These selected compounds are then subjected to preclinical experiments and clinical trials to assess their efficacy and safety. (Created with BioRender.com)

Disease modeling

For some rare and genetic diseases, the scarcity of clinical samples and the difficulty in establishing animal models have posed challenges to the study of their specific molecular mechanisms. USCs, as an excellent source of cells that can be non-invasively and conveniently obtained in large quantities from patients, can be reprogrammed into u-iPSCs and further differentiated into the cell types relevant for studying the diseases [ 29 ]. Many research teams have successfully utilized patient-derived u-iPSCs in conjunction with gene editing technologies to construct cell models for various human systems’ diseases. This serves as a powerful tool to help researchers better understand the disease mechanisms and develop new therapeutic strategies (Table  1 , see the end of the main text).

Muscular disease

Duchenne muscular dystrophy (DMD) is a hereditary muscle disease caused by mutations in the DMD gene on the X chromosome. DMD patients typically experience a lack or mutation in the DMD gene, preventing the normal production of the encoded muscle protein. This leads to damage in muscles such as skeletal, cardiac, and respisiratory muscles. Many DMD patients gradually develop movement disorders due to progressive muscle degeneration, and in severe cases, it can impact the heart and respiratory system, ultimately resulting in the patient’s death. Research indicates that, on average, four out of every five DMD patients succumbs to heart failure or respiratory failure [ 124 ]. To investigate this disease, Ghori et al. extracted urinary stem cells (USCs) from Pakistani children and efficiently reprogrammed them into u-iPSCs through transfection with episomal vectors [ 97 ]. Subsequently, after 11 days of in vitro induction, u-iPSCs successfully differentiated into DMD-Cardiomyocytes expressing cardiac markers such as NKX2-5 and TNNT-2 [ 62 ]. The establishment of this DMD cell model lays the foundation for further research into the molecular mechanisms of DMD and the identification of drug targets.

Ventricular Septal Defects (VSDs) is a common congenital heart disease. An abnormal opening in the ventricular septum allows the mixing of oxygenated and deoxygenated blood, causing the heart to pump blood more strenuously. This can ultimately lead to various complications, including symptoms such as shortness of breath, fatigue, and heart failure (HF) [ 125 ].

Cao et al. generated u-iPSCs with the ryanodine receptor type 2 (RyR2) mutation from a 2-month-old male patient with VSD with HF and directed the differentiation into functional cardiomyocytes by temporally manipulating canonical Wnt signaling using small molecules [ 61 ]. This study provides a robust cell model for investigation of the pathogenesis of VSD with HF.

Genitourinary disease

As cells derived from the kidney, USCs have been utilized in constructing models for kidney diseases [ 93 , 126 , 127 ]. X-linked Alport Syndrome (X-LAS), primarily caused by mutations in the gene encoding the protein COL4A5 in the renal tubular basement membrane, is an inheritable disorder affecting the renal tubular basement membrane. Damage to the renal tubular basement membrane leads to glomerulosclerosis and renal failure, resulting in clinical manifestations such as proteinuria, hematuria, and hypertension. Additionally, complications may involve the eyes and inner ears [ 128 , 129 ]. Guo et al. established a u-iPSC line with USCs harvested from a 5-year-old male X-LAS patient and demonstrated the feasibility as a cell-based disease models by verifying the expression of pluripotent makers, normal karyotype and capacity to differentiate into multiple germ layers [ 98 ].

A series of pediatric diseases are genetic in nature, including cystic fibrosis, congenital heart disease, spinal muscular atrophy, and others [ 130 , 131 , 132 ]. In the study of pediatric disease mechanisms, obstacles in obtaining early human embryos and associated ethical concerns have been significant limiting factors, highlighting the need for an appropriate in vitro research model. Cryptorchidism is a congenital reproductive system disorder characterized by the failure of the male testes to descend properly into the scrotum during embryonic development. Untreated cryptorchidism may lead to complications such as infertility and testicular cancer, posing reproductive health risks for affected individuals [ 133 ]. Zhou et al. reprogrammed USCs from a cryptorchid patient with mutations in genes including insulin-like factor 3 (INSL3), zinc finger (ZNF) 214, and ZNF215 into u-iPSCs [ 99 ]. By comparing them with human embryonic stem cells, the study confirmed their phenotypic, karyotypic, and pluripotent differentiation capabilities, providing a valuable in vitro model for understanding the disease mechanisms.

Blood disorder

Hemophilia is a common genetic bleeding disorder. The two most prevalent types are Hemophilia A and Hemophilia B, resulting from mutations in the F8 gene on the X chromosome and the F9 gene, respectively, leading to deficiencies in clotting factors VIII and IX [ 134 ]. Due to abnormalities in the clotting process, individuals with severe hemophilia may face life-threatening excessive bleeding during injuries or surgeries. The primary treatment involves replacing the deficient clotting factors by injecting plasma or preparations containing these factors [ 135 ]. However, one drawback is the development of antibodies, reducing the clinical effectiveness.

To advance gene therapy and novel clotting factor development, establishing appropriate disease models is crucial. Lu et al. generated u-iPSCs from a Hemophilia A patient with an Inv22 mutation through the electroporation of USCs using episomal plasmids [ 100 ]. Similarly, Ma et al. produced iPSCs from a Hemophilia B patient carrying the F9 variant c.223 C > T (p.R75X) [ 101 ]. The establishment of Hemophilia A and Hemophilia B iPSC lines serves as a robust tool for comprehending the underlying molecular mechanisms of hemophilia. In the studies by two other teams, they established Hemophilia A iPSC lines using USCs obtained from patient urine, and subsequently differentiated them in vitro into liver cells [ 67 ] and endothelial cells [ 102 ] with patient-specific mutations. Apart from hemophilia, the u-iPSC lines of another hemoglobin disorder, thalassemia, have been generated from patients carrying different mutations on globin genes [ 103 ]. The u-iPSC lines of various blood disorders provide a valuable cellular source for gene-corrected cell therapy.

Neurological disease

In the research of neurological disorders, the establishment of existing cell and animal disease models has provided powerful tools for studying the pathogenic mechanisms. However, the etiology of neurodevelopmental and neurodegenerative diseases is diverse, involving complex interactions between genetic and environmental factors that cannot be fully simulated in animal models. Some peripheral neuromuscular diseases may require sampling from patients, but obtaining samples of brain and spinal cord tissue is clinically challenging [ 136 ]. Therefore, finding suitable stem cells to establish disease models for in vitro dynamic and continuous research is crucial for understanding the occurrence and development of neurological disorders.

Neurodevelopmental disorders (NDDs) refer to defects or abnormalities in the early development of the central nervous system, leading to impaired functions such as behavior and cognition in patients. The manifestations of neurodevelopmental disorders often appear in preschool children, and their impact can last a lifetime, with most cases lacking clear treatment options [ 137 ]. U-iPSC lines have been established using urine from patients with NDDs, including autism spectrum disorder (ASD), developmental delay (DD), X-linked Renpenning syndrome (X-RSY), Down Syndrome, Attention Deficit Hyperactivity Disorder (ADHD), and TMC1-related hereditary deafness [ 34 , 104 , 105 , 106 , 107 ]. Teles et al. created three-dimensional human cerebral organoids with neurons and astrocytes differentiated from u-iPSCs derived from Down syndrome (DS) patients [ 34 ], which demonstrated developmental dynamics of the early-stage forebrain.

Neurodegenerative disorders (NDDs) involve the gradual degeneration of neurons in the brain and spinal cord, leading to irreversible cognitive impairments, motor dysfunction, and other symptoms [ 138 ]. Patient-specific u-iPSC lines for several neurodegenerative disorders, such as Alzheimer’s disease (AD) and Spinocerebellar ataxia type 3 (SCA3), have been developed in recent years. Sporadic Alzheimer’s disease (sAD), the most common form of dementia, predominantly influenced by genetic factors such as single nucleotide polymorphisms (SNPs). An iPSC line, KEIOi005-A, derived from USCs of a mild Alzheimer’s disease patient carrying multiple sporadic Alzheimer’s disease risk SNPs, exhibits normal stemness and pluripotency, and was suitable for in vitro modeling of sAD [ 108 ]. Spinocerebellar ataxia type 3 (SCA3), a neurodegenerative condition caused by a CAG repeat expansion in the ATXN3 gene, leads to progressive ataxia affecting balance, gait, and speech. The transformation of USCs from SCA3 patients into iPSCs hints at the potential of the ZZUi004-A iPSC line for studying SCA3’s underlying mechanisms, facilitating drug trials, and investigating gene therapy approaches [ 109 ].

In addition to NDDs, the establishment of u-iPSC lines for a movement disorder, paroxysmal kinesigenic dyskinesia (PKD), a tumor predisposition syndrome, neurofibromatosis type 1 (NF1), and brain tumor also demonstrates the value of u-iPSCs in modeling various neurological diseases [ 110 , 111 , 139 ]. Paroxysmal Kinesigenic Dyskinesia (PKD) is a genetic movement disorder linked to mutations in the PRRT2 gene. Disease-specific iPSCs generated from USCs from a PKD patient with a specific mutation present reduced PRRT2 expression and can differentiate into neurons. However, electrophysiological examinations find no significant differences compared to control cells. Overall, the study suggests that u-iPSCs offer a valuable tool for investigating PKD’s mechanisms [ 139 ]. Another study investigates using urine samples to generate iPSCs from pediatric brain tumor patients. These brain tumor iPSCs closely resemble iPSCs from non-tumor patients in terms of characteristics and ability to differentiate. Both types of iPSCs can efficiently turn into functional induced mesenchymal stem/stromal cells (iMSCs) with immunomodulatory properties, suggesting a promising non-invasive approach for personalized iMSC-based treatments for pediatric brain tumors [ 111 ].

Skeletal disorder

Musculoskeletal diseases rank second among global disabling conditions, imposing a significant burden on society [ 140 ]. As one of the most prevalent genetic diseases, hereditary musculoskeletal disorders can lead to fractures, muscle injuries, limited joint mobility, restricting patients’ daily activities, and diminishing their quality of life. In the exploration of the genetic factors underlying musculoskeletal disorders, researchers face challenges such as difficulties in obtaining samples from the human body and the lack of well-established models for rare diseases. The establishment of patient-derived iPSC lines and the in vitro directed osteogenesis provide suitable models for studying musculoskeletal diseases.

Several research teams have utilized u-iPSCs in the disease modeling of various genetic bone disorders, including Osteogenesis Imperfecta (OI) [ 112 ], Autosomal Dominant Osteopetrosis Type II (ADO2) [ 113 ], and Fibrodysplasia Ossificans Progressiva (FOP) [ 114 , 115 ]. OI is caused by mutations in collagen genes, resulting in bones that are prone to fractures, often accompanied by other connective tissue issues. Luan et al. generated an iPSC line from USCs of a 15-year-old female OI patient with a COL1A1 gene mutation using integration-free episomal vectors [ 112 ]. ADO2, a dominant inherited musculoskeletal disorder, leads to fractures, joint pain, and changes in bone morphology. Ou et al. produced ADO2-iPSCs from USCs of an ADO2 patient and identified the same CLCN7 mutation (R286W) present in the patient’s blood samples by comparing them with ADO2-iPSCs [ 113 ]. FOP is a rare disease caused by mutations in the ACVR1 gene, resulting in the gradual ossification of soft tissues, leading to loss of joint function and restricted movement. Two research groups generated iPSC lines with USCs from FOP patients carrying R206H mutations [ 114 , 115 ]. Cai et al. further directed the differentiation of FOP-iPSCs into endothelial cells and pericytes, revealing disease-related phenotypes in vitro [ 114 ].

Metabolic disorder

Inherited Metabolic Disorders (IMD) are caused by genetic mutations that result in structural and functional changes in the encoded protein molecules [ 141 ]. This leads to abnormalities in biochemical reactions and metabolism, with the accumulation of intermediate metabolites in the body, causing a range of clinical manifestations. IMDs have varied onset times, can affect multiple organs, and exhibit diverse clinical presentations [ 142 ]. Most IMDs currently lack effective treatments.

U-iPSCs have been used to establish disease models for various IMDs. Peroxisomes aid in the breakdown of fatty acids and hydrogen peroxide in the human body. When deficient, the accumulation of fatty acids and hydrogen peroxide can damage various tissues and organs. X-linked Adrenoleukodystrophy (X-ALD) is a hereditary metabolic disorder caused by peroxisomal dysfunction. It leads to progressive neurodegeneration, movement disorders, cognitive impairment, vision loss, adrenal insufficiency, and other symptoms. Wang et al. generated a u-iPSC line from a 6-year-old X-ALD patient with an ABCD1 gene mutation [ 116 ]. Additionally, IMDs related to amino acid metabolism, such as methylmalonic acidemia (MMA) and phenylketonuria (PKU), result from the deficiency of enzymes involved in their metabolism, leading to the accumulation of intermediate products. Han’s research group generated u-iPSC lines from a 10-year-old male MMA patient [ 117 ] and a 15-year-old male PKU patient [ 118 ]. Other u-iPSC lines for IMDs, including Barth syndrome and autosomal dominant hypercholesterolemia (ADH), have also been generated, providing a powerful tool for further understanding metabolic diseases [ 68 , 119 ].

Autoimmune disease

In autoimmune diseases, the immune system attacks the body’s own normal tissues, causing inflammation and damage to multiple organs throughout the body [ 143 ]. Autoimmune diseases often have multifaceted causes, including genetic, environmental, and immune system factors, making the design and implementation of clinical trials challenging. Establishing ideal animal or in vitro models and precisely regulating the immune system to minimize side effects are also challenges in research. Systemic Lupus Erythematosus (SLE) and Ankylosing Spondylitis (AS) are two distinct rheumatic diseases. The former can involve various organs throughout the body, causing symptoms such as rash, fatigue, and fever [ 144 ]. The latter primarily affects the spine and pelvic joints, manifesting as lower back pain and morning stiffness [ 145 ]. Chen et al. and Hu et al. generated u-iPSC lines from SLE patients and an AS patient with a JAK2 mutation, respectively, confirming USCs as an ideal source for modeling autoimmune diseases [ 120 , 121 ].

Retinal disorder

Inherited retinal diseases (IRDs) are a group of diseases characterized by progressive changes in the retina leading to vision loss, including X-linked Juvenile Retinoschisis (XLRS), Retinitis Pigmentosa (RP), and others. The global prevalence of monogenic IRDs is approximately 1 in 2,000, making them a significant cause of irreversible blindness in children and the working-age population [ 146 ]. Gene therapy is currently the only effective treatment for such diseases, and there is an urgent need for more experimental research and clinical trials to provide new therapeutic approaches. U-iPSCs have been used in modeling some inherited retinal diseases. Tang’s research group, for example, established two u-iPSC lines from individuals with specific conditions: an 11-year-old male with XLRS carrying a mutation in the retinoschisin gene (RS1) and a 17-year-old male patient with RP harboring a mutation in the pre-mRNA processing factor 8 gene (PRPF8) [ 122 , 123 ].

Epigenetic memory in u-iPSCs’ differentiation

Reports indicate that iPSCs derived from various somatic sources exhibit distinct epigenetic signatures, which influence their differentiation potential towards specific cell lineages associated with the donor tissue while hindering others. This “epigenetic memory” from the donor tissue may impede iPSC reprogramming efficiency and their ability to differentiate into desired cell types for disease modeling and treatment [ 147 ]. The impact of epigenetic memory on the differentiation of u-iPSCs is complex. Despite expressing mesenchymal stem cell (MSC) markers, u-iPSCs exhibit properties similar to parietal epithelial cells. While u-iPSCs can effectively differentiate into various cell lineages, they show a stronger propensity towards renal and epithelial cell types with tight junction and barrier function, indicating a nuanced view of epigenetic memory in u-iPSCs [ 148 , 149 ]. Although there are no observed negative effects on differentiation, further investigation into the underlying mechanisms and long-term consequences is needed.

Direct reprogramming reduces the risk of teratogenesis due to the lack of a pluripotent intermediate state and holds the potential of preserving the epigenetic memory of the donor cell, which has a tremendous impact on the accuracy of disease modeling [ 150 ]. By reprogramming USCs into iPSCs and subsequently directing the differentiation, the full course could extend to more than 12 weeks. Comparatively, through direct reprogramming, following expansion for 3 to 4 weeks, USCs undergo transduction with inducible MyoD (iMyoD) lentivirus, differentiate, and form myotubes in approximately 8 weeks. The shortening of culture time leads to reduced cost losses and increased efficiency, which offers an efficient and cost-effective method for generating patient-specific cell lineages [ 147 ].

Despite the advantages, optimization is required to enhance the efficiency of directed differentiation of USCs for generating target cells. Also, mature cells differentiated from reprogrammed USCs need thorough evaluation through genetic, biological, and functional assessments [ 151 ]. USCs exhibit potential for superior differentiation, making them valuable for studying mechanisms underlying both common and rare genetic diseases, as well as for drug screening purposes [ 152 ]. Future research should focus on understanding the specific epigenetic marks associated with different cell lineages, improving reprogramming techniques, optimizing lineage-specific differentiation protocols, and identifying pathways, growth factors, and culture conditions to overcome potential biases and enhance therapeutic applicability.

Challenges and considerations

Although u-iPSCs hold immense promise for precision medicine, several challenges and considerations must be addressed to fully exploit their potential. iPSCs exhibit high pluripotency, but their sensitivity to reprogramming varies depending on different cell sources, and their growth curves and differentiation tendencies may differ. Additionally, variations in reprogramming factor concentrations, types of transfection methods, cell culture conditions, and timing among different laboratories may lead to reduced iPSC induction efficiency and even the generation of off-target cells [ 153 ]. Therefore, establishing standardized reprogramming protocols is crucial to ensure the reproducibility and comparability of experiments across different research teams, which is key to improving the accuracy of experimental results. Standardization measures may include using the same cell source, selecting a set of standard classical reprogramming factors, maintaining consistency in experimental conditions, and establishing uniform iPSC identification criteria.

iPSCs have the ability for unlimited proliferation, but different cell lines have different mutation rates. Some genetic mutations may be introduced during iPSC reprogramming and amplification, leading to the occurrence of tumors [ 154 ]. Therefore, maintaining genomic stability of iPSCs during long-term expansion is crucial for ensuring their safety. Researchers conduct differentiation status checks, sequencing, and karyotype analysis on generated iPSCs to eliminate possible variations. Other methods include using non-integrative reprogramming methods, employing gene editing techniques to repair potential tumorigenic mutations in iPSCs, and pre-differentiating iPSCs into specific cell types in vitro [ 155 ]. In summary, stringent quality control and safety measures must be applied to iPSCs before clinical application to eliminate their tumorigenic potential.

In many current experimental results, iPSCs do express classical markers and possess specific morphological features. However, they may not function well in vivo. On one hand, iPSCs may aberrantly differentiate into teratomas, causing immune rejection. Moreover, the survival and engraftment of iPSCs in vivo require suitable conditions, and simple cell injections may not provide the appropriate microenvironment to promote their in vivo differentiation and maturation [ 156 ]. Therefore, researchers can consider a series of measures, including choosing the right treatment timing, an adequate number of cells, using biomaterials as scaffolds for iPSCs, and pre-differentiating them in vitro.

After addressing a series of laboratory issues, the goal is to achieve large-scale production of iPSCs to meet clinical needs. First, the selection of an appropriate cell source, such as USCs, which can be easily obtained in large quantities non-invasively, is crucial for large-scale expansion. Then, non-integrative reprogramming methods or gene editing techniques need to be adopted, along with optimized cell expansion strategies, and the establishment of scalable, automated culture systems to improve cell production efficiency. Additionally, regular testing and screening of iPSCs, timely removal of abnormal cells, and ensuring cell quality are essential. Finally, efficient purification and obtaining the desired cell types, along with the establishment of cell cryopreservation and recovery processes, are necessary for iPSCs to be used promptly when needed. Considering these steps collectively, pharmaceutical companies can mass-produce iPSCs, providing better tools for drug screening, disease modeling, and cell therapy [ 157 ].

U-iPSCs are a powerful tool for advancing precision medicine. Their unique advantages, such as non-invasive acquisition and high reprogramming efficiency, make them a promising source of patient-specific pluripotent cells for drug testing, disease modeling, and cell therapy. With further research and development, u-iPSCs have the potential to revolutionize the treatment of a wide range of diseases and improve healthcare outcomes for millions of patients. Overall, the review manuscript provides a comprehensive and insightful overview of the potential and application of u-iPSCs in precision medicine. It is evident that u-iPSCs are a powerful tool for advancing personalized healthcare and improving patient outcomes.

Availability of data and materials

Not applicable.

Solter D. From teratocarcinomas to embryonic stem cells and beyond: a history of embryonic stem cell research. Nat Rev Genet. 2006;7(4):319–27.

Article   CAS   PubMed   Google Scholar  

Shi L, Cui Y, Luan J, Zhou X, Han J. Urine-derived induced pluripotent stem cells as a modeling tool to study rare human diseases. Intractable Rare Dis Res. 2016;5(3):192–201.

Article   PubMed   PubMed Central   Google Scholar  

Jinek M, East A, Cheng A, Lin S, Ma E, Doudna J. RNA-programmed genome editing in human cells. Elife. 2013;2:e00471.

Cho SW, Kim S, Kim JM, Kim JS. Targeted genome engineering in human cells with the Cas9 RNA-guided endonuclease. Nat Biotechnol. 2013;31(3):230–2.

Bellin M, Marchetto MC, Gage FH, Mummery CL. Induced pluripotent stem cells: the new patient? Nat Rev Mol Cell Biol. 2012;13(11):713–26.

Article   PubMed   Google Scholar  

Streckfuss-Bömeke K, Wolf F, Azizian A, Stauske M, Tiburcy M, Wagner S, et al. Comparative study of human-induced pluripotent stem cells derived from bone marrow cells, hair keratinocytes, and skin fibroblasts. Eur Heart J. 2013;34(33):2618–29.

Staerk J, Dawlaty MM, Gao Q, Maetzel D, Hanna J, Sommer CA, et al. Reprogramming of human peripheral blood cells to induced pluripotent stem cells. Cell Stem Cell. 2010;7(1):20–4.

Article   CAS   PubMed   PubMed Central   Google Scholar  

Shi Y, Inoue H, Wu JC, Yamanaka S. Induced pluripotent stem cell technology: a decade of progress. Nat Rev Drug Discov. 2017;16(2):115–30.

Huang RL, Ma Q, Atala JX, Zhang AY. Body fluid-derived stem cells: an untapped stem cell source in genitourinary regeneration. Nat Rev Urol. 2023;20(12):739–61.

Wu R, Soland M, Liu G, Shi Y, Zhang C, Tang Y, et al. Functional characterization of the immunomodulatory properties of human urine-derived stem cells. Translational Androl Urol. 2021;10(9):3566–78.

Article   Google Scholar  

Zhou C, Wu X-R, Liu H-S, Liu X-H, Liu G-H, Zheng X-B, et al. Immunomodulatory Effect of urine-derived stem cells on inflammatory bowel diseases via Downregulating Th1/Th17 Immune responses in a PGE2-dependent manner. J Crohn’s Colitis. 2019;14(5):654–68.

Crisan M, Yap S, Casteilla L, Chen C-W, Corselli M, Park TS, et al. A perivascular origin for mesenchymal stem cells in multiple human organs. Cell Stem Cell. 2008;3(3):301–13.

Zhang Y, Atala A. Urothelial cell culture. Methods Mol Biol. 2013;1037:27–43.

Zhou T, Benda C, Duzinger S, Huang Y, Li X, Li Y, et al. Generation of induced pluripotent stem cells from urine. J Am Soc Nephrol. 2011;22(7):1221–8.

Liu Y, Zheng Y, Li S, Xue H, Schmitt K, Hergenroeder GW, et al. Human neural progenitors derived from integration-free iPSCs for SCI therapy. Stem cell Res. 2017;19:55–64.

Guan X, Mack DL, Moreno CM, Strande JL, Mathieu J, Shi Y, et al. Dystrophin-deficient cardiomyocytes derived from human urine: new biologic reagents for drug discovery. Stem Cell Res. 2014;12(2):467–80.

Benda C, Zhou T, Wang X, Tian W, Grillari J, Tse HF, et al. Urine as a source of stem cells. Adv Biochem Eng Biotechnol. 2013;129:19–32.

CAS   PubMed   Google Scholar  

Hockemeyer D, Soldner F, Beard C, Gao Q, Mitalipova M, DeKelver RC, et al. Efficient targeting of expressed and silent genes in human ESCs and iPSCs using zinc-finger nucleases. Nat Biotechnol. 2009;27(9):851–7.

Cong L, Ran FA, Cox D, Lin S, Barretto R, Habib N, et al. Multiplex genome engineering using CRISPR/Cas systems. Science. 2013;339(6121):819–23.

Lombardo A, Genovese P, Beausejour CM, Colleoni S, Lee YL, Kim KA, et al. Gene editing in human stem cells using zinc finger nucleases and integrase-defective lentiviral vector delivery. Nat Biotechnol. 2007;25(11):1298–306.

Urnov FD, Miller JC, Lee YL, Beausejour CM, Rock JM, Augustus S, et al. Highly efficient endogenous human gene correction using designed zinc-finger nucleases. Nature. 2005;435(7042):646–51.

Hockemeyer D, Wang H, Kiani S, Lai CS, Gao Q, Cassady JP, et al. Genetic engineering of human pluripotent cells using TALE nucleases. Nat Biotechnol. 2011;29(8):731–4.

Zou H, Guan M, Li Y, Luo F, Wang W, Qin Y. Targeted gene correction and functional recovery in achondroplasia patient-derived iPSCs. Stem Cell Res Ther. 2021;12(1):485.

Zhou M, Hu Z, Qiu L, Zhou T, Feng M, Hu Q, et al. Seamless Genetic Conversion of SMN2 to SMN1 via CRISPR/Cpf1 and single-stranded oligodeoxynucleotides in spinal muscular atrophy patient-specific Induced Pluripotent Stem cells. Hum Gene Ther. 2018;29(11):1252–63.

Neumeyer J, Lin R-Z, Wang K, Hong X, Hua T, Croteau SE, et al. Bioengineering hemophilia A–specific microvascular grafts for delivery of full-length factor VIII into the bloodstream. Blood Adv. 2019;3(24):4166–76.

Zeng B, Zhou M, Liu B, Shen F, Xiao R, Su J, et al. Targeted addition of mini-dystrophin into rDNA locus of Duchenne muscular dystrophy patient-derived iPSCs. Biochem Biophys Res Commun. 2021;545:40–5.

Ran FA, Hsu PD, Lin CY, Gootenberg JS, Konermann S, Trevino AE, et al. Double nicking by RNA-guided CRISPR Cas9 for enhanced genome editing specificity. Cell. 2013;154(6):1380–9.

Rossi G, Manfrin A, Lutolf MP. Progress and potential in organoid research. Nat Rev Genet. 2018;19(11):671–87.

Zhou T, Benda C, Dunzinger S, Huang Y, Ho JC, Yang J, et al. Generation of human induced pluripotent stem cells from urine samples. Nat Protoc. 2012;7(12):2080–9.

Mulder J, Sharmin S, Chow T, Rodrigues DC, Hildebrandt MR, D’Cruz R, et al. Generation of infant-and pediatric-derived urinary induced pluripotent stem cells competent to form kidney organoids. Pediatr Res. 2020;87(4):647–55.

Kim K, Gil M, Dayem A, Choi S, Kang G-H, Yang G-M, et al. Improved isolation and culture of urine-derived stem cells (USCs) and enhanced production of Immune cells from the USC-Derived Induced Pluripotent Stem cells. J Clin Med. 2020;9(3):827.

Nguyen L, Wruck W, Erichsen L, Graffmann N, Adjaye J. The nephrotoxin puromycin Aminonucleoside induces Injury in kidney organoids differentiated from Induced Pluripotent Stem cells. Cells. 2022;11(4):635.

Li G, Xie B, He L, Zhou T, Gao G, Liu S, et al. Generation of retinal organoids with mature rods and cones from urine-derived Human Induced Pluripotent Stem cells. Stem Cells Int. 2018;2018:1–12.

Google Scholar  

Teles ESAL, Yokota BY, Sertie AL, Zampieri BL. Generation of urine-derived Induced Pluripotent Stem cells and cerebral organoids for modeling Down Syndrome. Stem Cell Rev Rep. 2023;19(4):1116–23.

Cai J, Zhang Y, Liu P, Chen S, Wu X, Sun Y, et al. Generation of tooth-like structures from integration-free human urine induced pluripotent stem cells. Cell Regen. 2013;2(1):6.

Nguyen D, Hagg DA, Forsman A, Ekholm J, Nimkingratana P, Brantsing C, et al. Cartilage tissue Engineering by the 3D bioprinting of iPS cells in a Nanocellulose/Alginate Bioink. Sci Rep. 2017;7(1):658.

Varkey M, Visscher DO, van Zuijlen PPM, Atala A, Yoo JJ. Skin bioprinting: the future of burn wound reconstruction? Burns Trauma. 2019;7:4.

Ma X, Qu X, Zhu W, Li YS, Yuan S, Zhang H, et al. Deterministically patterned biomimetic human iPSC-derived hepatic model via rapid 3D bioprinting. Proc Natl Acad Sci U S A. 2016;113(8):2206–11.

Salaris F, Colosi C, Brighi C, Soloperto A, Turris V, Benedetti MC, et al. 3D Bioprinted Human cortical neural constructs derived from Induced Pluripotent Stem cells. J Clin Med. 2019;8(10):1595.

Yoshida Y, Yamanaka S. Induced Pluripotent stem cells 10 years later: for Cardiac Applications. Circ Res. 2017;120(12):1958–68.

Ortiz-Virumbrales M, Moreno CL, Kruglikov I, Marazuela P, Sproul A, Jacob S, et al. CRISPR/Cas9-Correctable mutation-related molecular and physiological phenotypes in iPSC-derived Alzheimer’s PSEN2 (N141I) neurons. Acta Neuropathol Commun. 2017;5(1):77.

Li H, Jiang H, Zhang B, Feng J. Modeling Parkinson’s Disease using patient-specific Induced Pluripotent Stem cells. J Parkinsons Dis. 2018;8(4):479–93.

Papapetrou EP. Patient-derived induced pluripotent stem cells in cancer research and precision oncology. Nat Med. 2016;22(12):1392–401.

Skardal A, Mack D, Kapetanovic E, Atala A, Jackson JD, Yoo J, et al. Bioprinted amniotic fluid-derived stem cells accelerate healing of large skin wounds. Stem Cells Transl Med. 2012;1(11):792–802.

Xing F, Yin H-M, Zhe M, Xie J-C, Duan X, Xu J-Z, et al. Nanotopographical 3D-Printed poly(ε-caprolactone) scaffolds enhance proliferation and osteogenic differentiation of urine-derived stem cells for bone regeneration. Pharmaceutics. 2022;14(7):1437.

Zhang X, Chen JL, Xing F, Duan X. Three-dimensional printed polylactic acid and hydroxyapatite composite scaffold with urine-derived stem cells as a treatment for bone defects. J Mater Sci Mater Med. 2022;33(10):71.

Shao R, Li C, Chen Y, Zhang L, Yang H, Zhang Z, et al. LncRNA-GAS5 promotes spinal cord repair and the inhibition of neuronal apoptosis via the transplantation of 3D printed scaffold loaded with induced pluripotent stem cell-derived neural stem cells. Annals Translational Med. 2021;9(11):931.

Article   CAS   Google Scholar  

Wagers AJ. The stem cell niche in regenerative medicine. Cell Stem Cell. 2012;10(4):362–9.

Ouyang B, Sun X, Han D, Chen S, Yao B, Gao Y, et al. Human urine-derived stem cells alone or genetically-modified with FGF2 improve type 2 diabetic erectile dysfunction in a rat model. PLoS ONE. 2014;9(3):e92825.

Dong X, Zhang T, Liu Q, Zhu J, Zhao J, Li J, et al. Beneficial effects of urine-derived stem cells on fibrosis and apoptosis of myocardial, glomerular and bladder cells. Mol Cell Endocrinol. 2016;427:21–32.

Zhang C, Luo D, Li T, Yang Q, Xie Y, Chen H, et al. Transplantation of human urine-derived stem cells ameliorates erectile function and cavernosal endothelial function by promoting autophagy of Corpus Cavernosal Endothelial Cells in Diabetic Erectile Dysfunction rats. Stem Cells Int. 2019;2019:2168709.

Ouyang B, Xie Y, Zhang C, Deng C, Lv L, Yao J, et al. Extracellular vesicles from human urine-derived stem cells ameliorate Erectile Dysfunction in a Diabetic Rat Model by delivering Proangiogenic MicroRNA. Sex Med. 2019;7(2):241–50.

Sun B, Dong X, Zhao J, Yang Z, Zhang Y, Li L. Differentiation of human urine-derived stem cells into interstitial cells of Cajal-like cells by exogenous gene modification: a preliminary study. Biochem Biophys Res Commun. 2020;523(1):10–7.

Li J, Luo H, Dong X, Liu Q, Wu C, Zhang T, et al. Therapeutic effect of urine-derived stem cells for protamine/lipopolysaccharide-induced interstitial cystitis in a rat model. Stem Cell Res Ther. 2017;8(1):107.

Zhang C, George SK, Wu R, Thakker PU, Abolbashari M, Kim TH, et al. Reno-protection of urine-derived stem cells in a chronic kidney Disease Rat Model Induced by Renal Ischemia and Nephrotoxicity. Int J Biol Sci. 2020;16(3):435–46.

Sun B, Luo X, Yang C, Liu P, Yang Y, Dong X, et al. Therapeutic effects of Human urine-derived stem cells in a rat model of Cisplatin-Induced Acute kidney Injury in vivo and in Vitro. Stem Cells Int. 2019;2019:8035076.

Deng C, Xie Y, Zhang C, Ouyang B, Chen H, Lv L, et al. Urine-derived stem cells facilitate endogenous spermatogenesis restoration of Busulfan-Induced Nonobstructive Azoospermic mice by Paracrine Exosomes. Stem Cells Dev. 2019;28(19):1322–33.

Young RA. Control of the embryonic stem cell state. Cell. 2011;144(6):940–54.

Ratajczak MZ, Zuba-Surma EK, Wysoczynski M, Wan W, Ratajczak J, Wojakowski W, et al. Hunt for pluripotent stem cell -- regenerative medicine search for almighty cell. J Autoimmun. 2008;30(3):151–62.

Wang C, Hei F, Ju Z, Yu J, Yang S, Chen M. Differentiation of urine-derived Human Induced Pluripotent Stem cells to alveolar type II epithelial cells. Cell Reprogram. 2016;18(1):30–6.

Cao Y, Xu J, Wen J, Ma X, Liu F, Li Y, et al. Generation of a urine-derived Ips cell line from a patient with a ventricular septal defect and heart failure and the robust differentiation of these cells to Cardiomyocytes via Small molecules. Cell Physiol Biochem. 2018;50(2):538–51.

Ghori FF, Wahid M. Induced pluripotent stem cells derived cardiomyocytes from Duchenne muscular dystrophy patients in vitro. Pak J Med Sci. 2021;37(5):1376–81.

Steinle H, Weber M, Behring A, Mau-Holzmann U, von Ohle C, Popov AF, et al. Reprogramming of urine-derived renal epithelial cells into iPSCs using srRNA and consecutive differentiation into beating cardiomyocytes. Mol Ther Nucleic Acids. 2019;17:907–21.

Kibschull M, Nguyen TTN, Chow T, Alarab M, Lye SJ, Rogers I, et al. Differentiation of patient-specific void urine-derived human induced pluripotent stem cells to fibroblasts and skeletal muscle myocytes. Sci Rep. 2023;13(1):4746.

Liu W, Zhang P, Tan J, Lin Y. Differentiation of urine-derived Induced Pluripotent Stem cells to neurons, astrocytes, and microvascular endothelial cells from a Diabetic patient. Cell Reprogram. 2020;22(3):147–55.

Liu A, Kang S, Yu P, Shi L, Zhou L. Transplantation of human urine-derived neural progenitor cells after spinal cord injury in rats. Neurosci Lett. 2020;735:135201.

Jia B, Chen S, Zhao Z, Liu P, Cai J, Qin D, et al. Modeling of hemophilia A using patient-specific induced pluripotent stem cells derived from urine cells. Life Sci. 2014;108(1):22–9.

Si-Tayeb K, Idriss S, Champon B, Caillaud A, Pichelin M, Arnaud L, et al. Urine-sample-derived human induced pluripotent stem cells as a model to study PCSK9-mediated autosomal dominant hypercholesterolemia. Dis Model Mech. 2016;9(1):81–90.

CAS   PubMed   PubMed Central   Google Scholar  

Fu Q, Qin Z, Jin X, Zhang L, Chen Z, He J, et al. Generation of Functional Lentoid bodies from Human Induced Pluripotent stem cells derived from urinary cells. Invest Opthalmology Visual Sci. 2017;58(1):517–27.

Jin Y, Zhang M, Li M, Zhang H, Zhang F, Zhang H, et al. Generation of urine-derived Induced Pluripotent Stem Cell line from patients with acute kidney Injury. Cell Reprogram. 2021;23(5):290–303.

Chawla LS, Eggers PW, Star RA, Kimmel PL. Acute kidney injury and chronic kidney disease as interconnected syndromes. N Engl J Med. 2014;371(1):58–66.

von Samson-Himmelstjerna FA, Kolbrink B, Schulte K. Relative excess mortality risk after kidney transplantation: Eve’s loss or Adam’s win? Kidney Int. 2023;104(3):619–20.

Gao WW, Chun SY, Kim BS, Ha YS, Lee JN, Lee EH, et al. Locally transplanted human urine-induced nephron progenitor cells contribute to renal repair in mice kidney with diabetic nephropathy. Biochem Biophys Res Commun. 2022;629:128–34.

Hampel C, Artibani W, Espuña Pons M, Haab F, Jackson S, Romero J, et al. Understanding the burden of stress urinary incontinence in Europe: a qualitative review of the literature. Eur Urol. 2004;46(1):15–27.

Ahuja CS, Nori S, Tetreault L, Wilson J, Kwon B, Harrop J, et al. Trauma Spinal Cord Injury-Repair Regeneration Neurosurg. 2017;80(3s):S9–22.

Thuret S, Moon LD, Gage FH. Therapeutic interventions after spinal cord injury. Nat Rev Neurosci. 2006;7(8):628–43.

Doi D, Magotani H, Kikuchi T, Ikeda M, Hiramatsu S, Yoshida K, et al. Pre-clinical study of induced pluripotent stem cell-derived dopaminergic progenitor cells for Parkinson’s disease. Nat Commun. 2020;11(1):3369.

Happle C, Lachmann N, Ackermann M, Mirenska A, Göhring G, Thomay K, et al. Pulmonary transplantation of Human Induced Pluripotent Stem Cell-derived macrophages ameliorates pulmonary alveolar proteinosis. Am J Respir Crit Care Med. 2018;198(3):350–60.

Shafa M, Ionescu LI, Vadivel A, Collins JJP, Xu L, Zhong S, et al. Human induced pluripotent stem cell-derived lung progenitor and alveolar epithelial cells attenuate hyperoxia-induced lung injury. Cytotherapy. 2018;20(1):108–25.

Ebert AD, Svendsen CN. Human stem cells and drug screening: opportunities and challenges. Nat Rev Drug Discov. 2010;9(5):367–72.

Vincent F, Loria P, Pregel M, Stanton R, Kitching L, Nocka K, et al. Developing predictive assays: the phenotypic screening rule of 3. Sci Transl Med. 2015;7(293):293ps15.

Moffat JG, Rudolph J, Bailey D. Phenotypic screening in cancer drug discovery - past, present and future. Nat Rev Drug Discov. 2014;13(8):588–602.

Inoue H, Nagata N, Kurokawa H, Yamanaka S. iPS cells: a game changer for future medicine. Embo j. 2014;33(5):409–17.

Li M, Zhao H, Ananiev GE, Musser MT, Ness KH, Maglaque DL, et al. Establishment of reporter lines for detecting Fragile X Mental Retardation (FMR1) gene reactivation in human neural cells. Stem Cells. 2017;35(1):158–69.

Kumari D, Swaroop M, Southall N, Huang W, Zheng W, Usdin K. High-throughput screening to identify compounds that increase Fragile X Mental retardation protein expression in neural stem cells differentiated from fragile X syndrome patient-derived Induced Pluripotent Stem cells. Stem Cells Transl Med. 2015;4(7):800–8.

Kaufmann M, Schuffenhauer A, Fruh I, Klein J, Thiemeyer A, Rigo P, et al. High-throughput screening using iPSC-Derived neuronal progenitors to identify compounds counteracting epigenetic gene silencing in Fragile X Syndrome. J Biomol Screen. 2015;20(9):1101–11.

Davis-Ajami ML, Fink JC, Wu J. Nephrotoxic medication exposure in U.S. adults with Predialysis chronic kidney disease: Health services utilization and cost outcomes. J Manag Care Spec Pharm. 2016;22(8):959–68.

PubMed   Google Scholar  

Chen N, Chen X, Ding X, Teng J. Analysis of the high incidence of acute kidney injury associated with acute-on-chronic liver failure. Hepatol Int. 2018;12(3):262–8.

Perazella MA, Rosner MH. Drug-Induced Acute kidney Injury. Clin J Am Soc Nephrol. 2022;17(8):1220–33.

Langhans SA. Three-Dimensional in Vitro Cell Culture models in Drug Discovery and Drug Repositioning. Front Pharmacol. 2018;9:6.

Guo H, Deng N, Dou L, Ding H, Criswell T, Atala A, et al. 3-D human renal tubular Organoids generated from urine-derived stem cells for Nephrotoxicity Screening. ACS Biomater Sci Eng. 2020;6(12):6701–9.

Sun G, Ding B, Wan M, Chen L, Jackson J, Atala A. Formation and optimization of three-dimensional organoids generated from urine-derived stem cells for renal function in vitro. Stem Cell Res Ther. 2020;11(1):309.

Takasato M, Er PX, Chiu HS, Maier B, Baillie GJ, Ferguson C, et al. Kidney organoids from human iPS cells contain multiple lineages and model human nephrogenesis. Nature. 2015;526(7574):564–8.

Dykens JA, Will Y. The significance of mitochondrial toxicity testing in drug development. Drug Discov Today. 2007;12(17–18):777–85.

Ding H, Jambunathan K, Jiang G, Margolis DM, Leng I, Ihnat M, et al. 3D spheroids of human primary urine-derived stem cells in the Assessment of Drug-Induced mitochondrial toxicity. Pharmaceutics. 2022;14(5):1042.

Ding H, George S, Leng XI, Ihnat M, Ma JX, Jiang G, et al. Silk fibers assisted long-term 3D culture of human primary urinary stem cells via inhibition of senescence-associated genes: potential use in the assessment of chronic mitochondrial toxicity. Mater Today Adv. 2022;15:100261.

Ghori FF, Wahid M. Induced pluripotent stem cells from urine of Duchenne muscular dystrophy patients. Pediatr Int. 2021;63(9):1038–47.

Guo X, Ji W, Niu C, Ding Y, Chen Z, Chen C, et al. Generation of an urine-derived induced pluripotent stem cell line from a 5-year old X-linked Alport syndrome (X-LAS) patient. Stem cell Res. 2020;49:102085.

Zhou J, Wang X, Zhang S, Gu Y, Yu L, Wu J, et al. Generation and characterization of human cryptorchid-specific induced pluripotent stem cells from urine. Stem Cells Dev. 2013;22(5):717–25.

Lu D, Xue Y, Song B, Liu N, Xie Y, Cheng Y, et al. Generation of induced pluripotent stem cell GZLSL-i001-A derived from urine-derived cells of Hemophilia A patient with Inv22 mutation. Stem cell Res. 2020;49:102053.

Ma Y, Sun W, Liu X, Ren J, Zhang X, Zhang R, et al. Generation an induced pluripotent stem cell line SXMUi001-A derived from a hemophilia B patient carries variant F9 c.223C>T(p.R75X). Stem Cell Res. 2022;60:102684.

Hu Z, Zhou M, Wu Y, Li Z, Liu X, Wu L, et al. ssODN-Mediated In-Frame deletion with CRISPR/Cas9 restores FVIII function in Hemophilia a-patient-derived iPSCs and ECs. Mol Ther Nucleic Acids. 2019;17:198–209.

Long P, Wang Z, Yang H, Liu Z, Wu B, Zhong G, et al. Generation of nine iPSC lines (HNMUi002-A, HNMUi003-A, HNMUi004-A, HNMUi005-A, HNMUi006-A, HNMUi007-A, HNMUi008-A, HNMUi009-A, HNMUi010-A) from three Chinese families with thalassemia. Stem cell Res. 2020;49:102014.

Yang X, Zhou T, Zhang H, Li Y, Dong R, Liu N, et al. Generation of an induced pluripotent stem cell line (SDQLCHi008-A) from a patient with ASD and DD carrying an 830 kb de novo deletion at chr7q11.22 including the exon 1 of AUTS2 gene. Stem Cell Res. 2019;40:101557.

Guo X, Zhang X, Wang L, He L, Ding Y, Chen H, et al. Generation of an urine-derived induced pluripotent stem cell line WMUi017-A from a X-linked renpenning syndrome (X-RSY) patient with the hemizygous PQBP1 gene mutation p.P609A (c.1825C > G). Stem Cell Res. 2021;51:102159.

Wang H, Wu K, Guan J, Wang Q. Generation of a human induced pluripotent stem cell line (CPGHi001-A) from a hearing loss patient with the TMC1 p.M418K mutation. Stem Cell Res. 2020;49:101982.

Sochacki J, Devalle S, Reis M, Mattos P, Rehen S. Generation of urine iPS cell lines from patients with attention deficit hyperactivity disorder (ADHD) using a non-integrative method. Stem Cell Res. 2016;17(1):102–6.

Supakul S, Leventoux N, Tabuchi H, Mimura M, Ito D, Maeda S, et al. Establishment of KEIOi005-A iPSC line from urine-derived cells (UDCs) of a mild Alzheimer’s disease (AD) donor with multiple risk SNPs for sporadic Alzheimer’s disease (sAD). Stem Cell Res. 2022;62:102802.

Wang Y, Shi C, Wang Z, Sun H, Yang Z, Zhang F, et al. Generation of induced pluripotent stem cell line (ZZUi004-A) from urine sample of a patient with spinocerebellar ataxia type 3. Stem Cell Res. 2018;28:71–4.

Shi L, Cui Y, Qi Z, Zhou X, Luan J, Han J. Generation of two non-integrated induced pluripotent stem cell lines from urine-derived cells of a Chinese patient carrying NF1 gene mutation. Stem Cell Res. 2020;46:101842.

Baliña-Sánchez C, Aguilera Y, Adán N, Sierra-Párraga JM, Olmedo-Moreno L, Panadero-Morón C, et al. Generation of mesenchymal stromal cells from urine-derived iPSCs of pediatric brain tumor patients. Front Immunol. 2023;14:1022676.

Luan J, Cui Y, Wang J, Liang Y, Zhao Y, Zhang G, et al. Generation of a non-integrated induced pluripotent stem cell line from urine cells of a Chinese osteogenesis imperfecta type I patient. Stem Cell Res. 2022;62:102827.

Ou M, Li C, Tang D, Xue W, Xu Y, Zhu P, et al. Genotyping, generation and proteomic profiling of the first human autosomal dominant osteopetrosis type II-specific induced pluripotent stem cells. Stem Cell Res Ther. 2019;10(1):251.

Cai J, Orlova VV, Cai X, Eekhoff EMW, Zhang K, Pei D, et al. Induced Pluripotent Stem cells to Model Human Fibrodysplasia ossificans Progressiva. Stem Cell Rep. 2015;5(6):963–70.

Hildebrand L, Rossbach B, Kuhnen P, Gossen M, Kurtz A, Reinke P, et al. Generation of integration free induced pluripotent stem cells from fibrodysplasia ossificans progressiva (FOP) patients from urine samples. Stem Cell Res. 2016;16(1):54–8.

Wang L, Gao B, Mo X, Guo X, Huang J. Generation of an urine-derived induced pluripotent stem cell line from a 6-year old X-linked adrenoleukodystrophy (X-ALD) patient. Stem Cell Res. 2021;51:102170.

Luan J, Zou H, Cui Y, Wang J, Han Z, Zhang G, et al. Generation of induced pluripotent stem cells named SMBCi019-A from a methylmalonic acidemia patient carrying the MMACHC mutations. Stem Cell Res. 2022;62:102821.

Qi Z, Cui Y, Shi L, Luan J, Zhou X, Han J. Generation of urine-derived induced pluripotent stem cells from a patient with phenylketonuria. Intractable Rare Dis Res. 2018;7(2):87–93.

Guo X, Wang L, Chen K, Song S, Wang X, Gu X, et al. Generation of urine-derived iPS cell line via a non-integrative method from a Barth syndrome patient with TAZ gene mutation. Stem cell Res. 2020;47:101886.

Chen Y, Luo R, Xu Y, Cai X, Li W, Tan K, et al. Generation of systemic lupus erythematosus-specific induced pluripotent stem cells from urine. Rheumatol Int. 2013;33(8):2127–34.

Hu J, Ren W, Qiu W, Lv J, Zhang C, Xu C, et al. Generation of induced pluripotent stem cell line (XDCMHi001-A) from an ankylosing spondylitis patient with JAK2 mutation. Stem cell Res. 2020;45:101788.

Yan X, Guo Y, Chen J, Cui Z, Gu J, Wang Y, et al. Establishment of CSUASOi001-A, a non-integrated induced pluripotent stem cell line from urine-derived cells of a Chinese patient carrying RS1 gene mutation. Stem cell Res. 2019;38:101466.

Zhou Y, Cui Z, Jing Y, Mao S, Chen D, Ding C, et al. Establishment of non-integrate induced pluripotent stem cell line CSUASOi006-A, from urine-derived cells of a PRPF8-related dominant retinitis pigmentosa patient. Stem cell Res. 2020;49:102041.

Wahlgren L, Kroksmark AK, Tulinius M, Sofou K. One in five patients with Duchenne muscular dystrophy dies from other causes than cardiac or respiratory failure. Eur J Epidemiol. 2022;37(2):147–56.

Penny DJ, Vick GW. 3rd. Ventricular septal defect. Lancet. 2011;377(9771):1103–12.

Lazzeri E, Ronconi E, Angelotti ML, Peired A, Mazzinghi B, Becherucci F, et al. Human urine-derived renal progenitors for Personalized modeling of genetic kidney disorders. J Am Soc Nephrology: JASN. 2015;26(8):1961–74.

Yin X, Li Q, McNutt PM, Zhang Y. Urine-derived stem cells for epithelial tissues Reconstruction and Wound Healing. Pharmaceutics. 2022;14(8):1669.

Jais JP, Knebelmann B, Giatras I, De Marchi M, Rizzoni G, Renieri A, et al. X-linked Alport syndrome: natural history and genotype-phenotype correlations in girls and women belonging to 195 families: a European Community Alport Syndrome Concerted Action study. J Am Soc Nephrol. 2003;14(10):2603–10.

Knebelmann B, Breillat C, Forestier L, Arrondel C, Jacassier D, Giatras I, et al. Spectrum of mutations in the COL4A5 collagen gene in X-linked Alport syndrome. Am J Hum Genet. 1996;59(6):1221–32.

Mercuri E, Bertini E, Iannaccone ST. Childhood spinal muscular atrophy: controversies and challenges. Lancet Neurol. 2012;11(5):443–52.

Grasemann H, Ratjen F. Early lung disease in cystic fibrosis. Lancet Respir Med. 2013;1(2):148–57.

Hinton RB, Ware SM. Heart failure in Pediatric patients with congenital heart disease. Circ Res. 2017;120(6):978–94.

Virtanen HE, Toppari J. Epidemiology and pathogenesis of cryptorchidism. Hum Reprod Update. 2008;14(1):49–58.

Soucie JM, Evatt B, Jackson D. Occurrence of hemophilia in the United States. Am J Hematol. 1998;59(4):288–94.

Mannucci PM. Hemophilia therapy: the future has begun. Haematologica. 2020;105(3):545–53.

Han SS, Williams LA, Eggan KC. Constructing and deconstructing stem cell models of neurological disease. Neuron. 2011;70(4):626–44.

Thapar A, Cooper M, Rutter M. Neurodevelopmental disorders. Lancet Psychiatry. 2017;4(4):339–46.

Rachakonda V, Pan TH, Le WD. Biomarkers of neurodegenerative disorders: how good are they? Cell Res. 2004;14(5):347–58.

Zhang SZ, Li HF, Ma LX, Qian WJ, Wang ZF, Wu ZY. Urine-derived induced pluripotent stem cells as a modeling tool for paroxysmal kinesigenic dyskinesia. Biol Open. 2015;4(12):1744–52.

Vos T, Flaxman AD, Naghavi M, Lozano R, Michaud C, Ezzati M, et al. Years lived with disability (YLDs) for 1160 sequelae of 289 diseases and injuries 1990–2010: a systematic analysis for the global burden of Disease Study 2010. Lancet. 2012;380(9859):2163–96.

Wolf B. Biotinidase deficiency: if you have to have an inherited metabolic disease, this is the one to have. Genet Med. 2012;14(6):565–75.

Michalski A, Leonard JV, Taylor DS. The eye and inherited metabolic disease: a review. J R Soc Med. 1988;81(5):286–90.

Wang L, Wang FS, Gershwin ME. Human autoimmune diseases: a comprehensive update. J Intern Med. 2015;278(4):369–95.

Fanouriakis A, Kostopoulou M, Alunno A, Aringer M, Bajema I, Boletis JN, et al. 2019 update of the EULAR recommendations for the management of systemic lupus erythematosus. Ann Rheum Dis. 2019;78(6):736–45.

Braun J, Sieper J. Ankylosing spondylitis. Lancet. 2007;369(9570):1379–90.

Berger W, Kloeckener-Gruissem B, Neidhardt J. The molecular basis of human retinal and vitreoretinal diseases. Prog Retin Eye Res. 2010;29(5):335–75.

Kim EY, Page P, Dellefave-Castillo LM, McNally EM, Wyatt EJ. Direct reprogramming of urine-derived cells with inducible MyoD for modeling human muscle disease. Skelet Muscle. 2016;6(1):32.

Polo JM, Liu S, Figueroa ME, Kulalert W, Eminli S, Tan KY, et al. Cell type of origin influences the molecular and functional properties of mouse induced pluripotent stem cells. Nat Biotechnol. 2010;28(8):848–55.

Rouhani F, Kumasaka N, de Brito MC, Bradley A, Vallier L, Gaffney D. Genetic background drives transcriptional variation in human induced pluripotent stem cells. PLoS Genet. 2014;10(6):e1004432.

Sato M, Takizawa H, Nakamura A, Turner BJ, Shabanpoor F, Aoki Y. Application of urine-derived stem cells to Cellular modeling in neuromuscular and neurodegenerative diseases. Front Mol Neurosci. 2019;12:297.

Scesa G, Adami R, Bottai D. iPSC Preparation and Epigenetic Memory: does the tissue origin. Matter? Cells. 2021;10(6):1470.

Poetsch MS, Strano A, Guan K. Human induced pluripotent stem cells: from cell origin, genomic stability, and epigenetic memory to translational medicine. Stem Cells. 2022;40(6):546–55.

Maza I, Caspi I, Zviran A, Chomsky E, Rais Y, Viukov S, et al. Transient acquisition of pluripotency during somatic cell transdifferentiation with iPSC reprogramming factors. Nat Biotechnol. 2015;33(7):769–74.

Lund RJ, Närvä E, Lahesmaa R. Genetic and epigenetic stability of human pluripotent stem cells. Nat Rev Genet. 2012;13(10):732–44.

Martin U. Genome stability of programmed stem cell products. Adv Drug Deliv Rev. 2017;120:108–17.

Fu X, Xu Y. Challenges to the clinical application of pluripotent stem cells: towards genomic and functional stability. Genome Med. 2012;4(6):55.

Lu X, Zhao T. Clinical therapy using iPSCs: hopes and challenges. Genomics Proteom Bioinf. 2013;11(5):294–8.

Download references

Acknowledgements

The authors would like to acknowledge funding supports Federal funds from the National Institute of Allergy and Infectious Diseases, and National Eye Institute National Institutes of Health, under Contract No. R21 AI152832, R03 AI165170 and R21EY035833.(PI: Y. Z).

Author information

Authors and affiliations.

Department of Plastic and Reconstructive Surgery, Shanghai Ninth People’s Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China

Xiya Yin & Qingfeng Li

Department of Burn and Plastic Surgery, West China Hospital, Sichuan University, Chengdu, China

Department of Pharmaceutical Sciences, School of Pharmacy, University of Maryland, Baltimore, Baltimore, MD, USA

Yan Shu & Hongbing Wang

Keck School of Medicine, Roski Eye Institute, University of Southern California, Los Angeles, CA, 90033, USA

Biju Thomas

Wake Forest Institute for Regenerative Medicine, Wake Forest University Health Sciences, Winston-Salem, NC, USA

Joshua T. Maxwell & Yuanyuan Zhang

You can also search for this author in PubMed   Google Scholar

Contributions

Xiya Yin: Conceptualization, Investigation, Writing original draft, Review and editing, Supervision. Qingfeng Li: Conceptualization, Supervision, Project administration, Funding acquisition. Yan Shu: Conceptualization, Formal analysis, and Review. Hongbing Wang: Investigation, visualization, and Resources. Biju Thomas: Writing original draft, Review, and editing. Joshua T. Maxwell: Review, and revision. Yuanyuan Zhang: Conceptualization, Review and editing, Supervision, Project administration, Funding acquisition.

Corresponding authors

Correspondence to Qingfeng Li or Yuanyuan Zhang .

Ethics declarations

Ethics approval and consent to participate, consent for publication, competing interests.

The authors declare that they have no competing interests.

Additional information

Publisher’s note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ . The Creative Commons Public Domain Dedication waiver ( http://creativecommons.org/publicdomain/zero/1.0/ ) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Cite this article.

Yin, X., Li, Q., Shu, Y. et al. Exploiting urine-derived induced pluripotent stem cells for advancing precision medicine in cell therapy, disease modeling, and drug testing. J Biomed Sci 31 , 47 (2024). https://doi.org/10.1186/s12929-024-01035-4

Download citation

Received : 26 January 2024

Accepted : 30 April 2024

Published : 09 May 2024

DOI : https://doi.org/10.1186/s12929-024-01035-4

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Urine-derived stem cell
• Disease modelling
• Regenerative medicine
• Personalized medicine

Journal of Biomedical Science

ISSN: 1423-0127

• Submission enquiries: Access here and click Contact Us
• General enquiries: [email protected]

We've detected unusual activity from your computer network

To continue, please click the box below to let us know you're not a robot.

Why did this happen?

Please make sure your browser supports JavaScript and cookies and that you are not blocking them from loading. For more information you can review our Terms of Service and Cookie Policy .

For inquiries related to this message please contact our support team and provide the reference ID below.

AlphaFold 3 predicts the structure and interactions of all of life’s molecules

May 08, 2024

[[read-time]] min read

Introducing AlphaFold 3, a new AI model developed by Google DeepMind and Isomorphic Labs. By accurately predicting the structure of proteins, DNA, RNA, ligands and more, and how they interact, we hope it will transform our understanding of the biological world and drug discovery.

Inside every plant, animal and human cell are billions of molecular machines. They’re made up of proteins, DNA and other molecules, but no single piece works on its own. Only by seeing how they interact together, across millions of types of combinations, can we start to truly understand life’s processes.

In a paper published in Nature , we introduce AlphaFold 3, a revolutionary model that can predict the structure and interactions of all life’s molecules with unprecedented accuracy. For the interactions of proteins with other molecule types we see at least a 50% improvement compared with existing prediction methods, and for some important categories of interaction we have doubled prediction accuracy.

We hope AlphaFold 3 will help transform our understanding of the biological world and drug discovery. Scientists can access the majority of its capabilities, for free, through our newly launched AlphaFold Server , an easy-to-use research tool. To build on AlphaFold 3’s potential for drug design, Isomorphic Labs is already collaborating with pharmaceutical companies to apply it to real-world drug design challenges and, ultimately, develop new life-changing treatments for patients.

Our new model builds on the foundations of AlphaFold 2, which in 2020 made a fundamental breakthrough in protein structure prediction . So far, millions of researchers globally have used AlphaFold 2 to make discoveries in areas including malaria vaccines, cancer treatments and enzyme design. AlphaFold has been cited more than 20,000 times and its scientific impact recognized through many prizes, most recently the Breakthrough Prize in Life Sciences . AlphaFold 3 takes us beyond proteins to a broad spectrum of biomolecules. This leap could unlock more transformative science, from developing biorenewable materials and more resilient crops, to accelerating drug design and genomics research.

7PNM - Spike protein of a common cold virus (Coronavirus OC43): AlphaFold 3’s structural prediction for a spike protein (blue) of a cold virus as it interacts with antibodies (turquoise) and simple sugars (yellow), accurately matches the true structure (gray). The animation shows the protein interacting with an antibody, then a sugar. Advancing our knowledge of such immune-system processes helps better understand coronaviruses, including COVID-19, raising possibilities for improved treatments.

How AlphaFold 3 reveals life’s molecules

Given an input list of molecules, AlphaFold 3 generates their joint 3D structure, revealing how they all fit together. It models large biomolecules such as proteins, DNA and RNA, as well as small molecules, also known as ligands — a category encompassing many drugs. Furthermore, AlphaFold 3 can model chemical modifications to these molecules which control the healthy functioning of cells, that when disrupted can lead to disease.

AlphaFold 3’s capabilities come from its next-generation architecture and training that now covers all of life’s molecules. At the core of the model is an improved version of our Evoformer module — a deep learning architecture that underpinned AlphaFold 2’s incredible performance. After processing the inputs, AlphaFold 3 assembles its predictions using a diffusion network, akin to those found in AI image generators. The diffusion process starts with a cloud of atoms, and over many steps converges on its final, most accurate molecular structure.

AlphaFold 3’s predictions of molecular interactions surpass the accuracy of all existing systems. As a single model that computes entire molecular complexes in a holistic way, it’s uniquely able to unify scientific insights.

7R6R - DNA binding protein: AlphaFold 3’s prediction for a molecular complex featuring a protein (blue) bound to a double helix of DNA (pink) is a near-perfect match to the true molecular structure discovered through painstaking experiments (gray).

Leading drug discovery at Isomorphic Labs

AlphaFold 3 creates capabilities for drug design with predictions for molecules commonly used in drugs, such as ligands and antibodies, that bind to proteins to change how they interact in human health and disease.

AlphaFold 3 achieves unprecedented accuracy in predicting drug-like interactions, including the binding of proteins with ligands and antibodies with their target proteins. AlphaFold 3 is 50% more accurate than the best traditional methods on the PoseBusters benchmark without needing the input of any structural information, making AlphaFold 3 the first AI system to surpass physics-based tools for biomolecular structure prediction. The ability to predict antibody-protein binding is critical to understanding aspects of the human immune response and the design of new antibodies — a growing class of therapeutics.

Using AlphaFold 3 in combination with a complementary suite of in-house AI models, Isomorphic Labs is working on drug design for internal projects as well as with pharmaceutical partners. Isomorphic Labs is using AlphaFold 3 to accelerate and improve the success of drug design — by helping understand how to approach new disease targets, and developing novel ways to pursue existing ones that were previously out of reach.

AlphaFold Server: A free and easy-to-use research tool

8AW3 - RNA modifying protein: AlphaFold 3’s prediction for a molecular complex featuring a protein (blue), a strand of RNA (purple), and two ions (yellow) closely matches the true structure (gray). This complex is involved with the creation of other proteins — a cellular process fundamental to life and health.

Google DeepMind’s newly launched AlphaFold Server is the most accurate tool in the world for predicting how proteins interact with other molecules throughout the cell. It is a free platform that scientists around the world can use for non-commercial research. With just a few clicks, biologists can harness the power of AlphaFold 3 to model structures composed of proteins, DNA, RNA and a selection of ligands, ions and chemical modifications.

AlphaFold Server helps scientists make novel hypotheses to test in the lab, speeding up workflows and enabling further innovation. Our platform gives researchers an accessible way to generate predictions, regardless of their access to computational resources or their expertise in machine learning.

Experimental protein-structure prediction can take about the length of a PhD and cost hundreds of thousands of dollars. Our previous model, AlphaFold 2, has been used to predict hundreds of millions of structures, which would have taken hundreds of millions of researcher-years at the current rate of experimental structural biology.

Sharing the power of AlphaFold 3 responsibly

With each AlphaFold release, we’ve sought to understand the broad impact of the technology , working together with the research and safety community. We take a science-led approach and have conducted extensive assessments to mitigate potential risks and share the widespread benefits to biology and humanity.

Building on the external consultations we carried out for AlphaFold 2, we’ve now engaged with more than 50 domain experts, in addition to specialist third parties, across biosecurity, research and industry, to understand the capabilities of successive AlphaFold models and any potential risks. We also participated in community-wide forums and discussions ahead of AlphaFold 3’s launch.

AlphaFold Server reflects our ongoing commitment to share the benefits of AlphaFold, including our free database of 200 million protein structures. We’ll also be expanding our free AlphaFold education online course with EMBL-EBI and partnerships with organizations in the Global South to equip scientists with the tools they need to accelerate adoption and research, including on underfunded areas such as neglected diseases and food security. We’ll continue to work with the scientific community and policy makers to develop and deploy AI technologies responsibly.

Opening up the future of AI-powered cell biology

7BBV - Enzyme: AlphaFold 3’s prediction for a molecular complex featuring an enzyme protein (blue), an ion (yellow sphere) and simple sugars (yellow), along with the true structure (gray). This enzyme is found in a soil-borne fungus (Verticillium dahliae) that damages a wide range of plants. Insights into how this enzyme interacts with plant cells could help researchers develop healthier, more resilient crops.

AlphaFold 3 brings the biological world into high definition. It allows scientists to see cellular systems in all their complexity, across structures, interactions and modifications. This new window on the molecules of life reveals how they’re all connected and helps understand how those connections affect biological functions — such as the actions of drugs, the production of hormones and the health-preserving process of DNA repair.

The impacts of AlphaFold 3 and our free AlphaFold Server will be realized through how they empower scientists to accelerate discovery across open questions in biology and new lines of research. We’re just beginning to tap into AlphaFold 3’s potential and can’t wait to see what the future holds.

Related stories

Bringing Project Starline out of the lab

6 incredible images of the human brain built with the help of Google's AI

How Prados Beauty is using Gemini to grow their business

3 things we learned from professional creatives about their hopes for AI

A new report explores the economic impact of generative AI

Enhance visual storytelling in Demand Gen with generative AI

Let’s stay in touch. Get the latest news from Google in your inbox.

Suggestions or feedback?

MIT News | Massachusetts Institute of Technology

• Machine learning
• Social justice
• Black holes
• Classes and programs

Departments

• Aeronautics and Astronautics
• Brain and Cognitive Sciences
• Architecture
• Political Science
• Mechanical Engineering

Centers, Labs, & Programs

• Abdul Latif Jameel Poverty Action Lab (J-PAL)
• Picower Institute for Learning and Memory
• Lincoln Laboratory
• School of Architecture + Planning
• School of Engineering
• School of Humanities, Arts, and Social Sciences
• Sloan School of Management
• School of Science
• MIT Schwarzman College of Computing

New treatment could reverse hair loss caused by an autoimmune skin disease

Press contact :, media download.

*Terms of Use:

Images for download on the MIT News office website are made available to non-commercial entities, press and the general public under a Creative Commons Attribution Non-Commercial No Derivatives license . You may not alter the images provided, other than to crop them to size. A credit line must be used when reproducing images; if one is not provided below, credit the images to "MIT."

Previous image Next image

Researchers at MIT, Brigham and Women’s Hospital, and Harvard Medical School have developed a potential new treatment for alopecia areata, an autoimmune disorder that causes hair loss and affects people of all ages, including children.

For most patients with this type of hair loss, there is no effective treatment. The team developed a microneedle patch that can be painlessly applied to the scalp and releases drugs that help to rebalance the immune response at the site, halting the autoimmune attack.

In a study of mice, the researchers found that this treatment allowed hair to regrow and dramatically reduced inflammation at the treatment site, while avoiding systemic immune effects elsewhere in the body. This strategy could also be adapted to treat other autoimmune skin diseases such as vitiligo, atopic dermatitis, and psoriasis, the researchers say.

“This innovative approach marks a paradigm shift. Rather than suppressing the immune system, we’re now focusing on regulating it precisely at the site of antigen encounter to generate immune tolerance,” says Natalie Artzi, a principal research scientist in MIT’s Institute for Medical Engineering and Science, an associate professor of medicine at Harvard Medical School and Brigham and Women’s Hospital, and an associate faculty member at the Wyss Institute of Harvard University.

Artzi and Jamil R. Azzi, an associate professor of medicine at Harvard Medical School and Brigham and Women’s Hospital, are the senior authors of the new study , which appears in the journal Advanced Materials . Nour Younis, a Brigham and Women’s postdoc, and Nuria Puigmal, a Brigham and Women’s postdoc and former MIT research affiliate, are the lead authors of the paper.

The researchers are now working on launching a company to further develop the technology, led by Puigmal, who was recently awarded a Harvard Business School Blavatnik Fellowship.

Direct delivery

Alopecia areata, which affects more than 6 million Americans, occurs when the body’s own T cells attack hair follicles, leading the hair to fall out. The only treatment available to most patients — injections of immunosuppressant steroids into the scalp — is painful and patients often can’t tolerate it.

Some patients with alopecia areata and other autoimmune skin diseases can also be treated with immunosuppressant drugs that are given orally, but these drugs lead to widespread suppression of the immune system, which can have adverse side effects.

“This approach silences the entire immune system, offering relief from inflammation symptoms but leading to frequent recurrences. Moreover, it increases susceptibility to infections, cardiovascular diseases, and cancer,” Artzi says.

A few years ago, at a working group meeting in Washington, Artzi happened to be seated next to Azzi (the seating was alphabetical), an immunologist and transplant physican who was seeking new ways to deliver drugs directly to the skin to treat skin-related diseases.

Their conversation led to a new collaboration, and the two labs joined forces to work on a microneedle patch to deliver drugs to the skin. In 2021, they reported that such a patch can be used to prevent rejection following skin transplant. In the new study, they began applying this approach to autoimmune skin disorders.

“The skin is the only organ in our body that we can see and touch, and yet when it comes to drug delivery to the skin, we revert to systemic administration. We saw great potential in utilizing the microneedle patch to reprogram the immune system locally,” Azzi says.

The microneedle patches used in this study are made from hyaluronic acid crosslinked with polyethylene glycol (PEG), both of which are biocompatible and commonly used in medical applications. With this delivery method, drugs can pass through the tough outer layer of the epidermis, which can’t be penetrated by creams applied to the skin.

“This polymer formulation allows us to create highly durable needles capable of effectively penetrating the skin. Additionally, it gives us the flexibility to incorporate any desired drug,” Artzi says. For this study, the researchers loaded the patches with a combination of the cytokines IL-2 and CCL-22. Together, these immune molecules help to recruit regulatory T cells, which proliferate and help to tamp down inflammation. These cells also help the immune system learn to recognize that hair follicles are not foreign antigens, so that it will stop attacking them.

Hair regrowth

The researchers found that mice treated with this patch every other day for three weeks had many more regulatory T cells present at the site, along with a reduction in inflammation. Hair was able to regrow at those sites, and this growth was maintained for several weeks after the treatment ended. In these mice, there were no changes in the levels of regulatory T cells in the spleen or lymph nodes, suggesting that the treatment affected only the site where the patch was applied.

In another set of experiments, the researchers grafted human skin onto mice with a humanized immune system. In these mice, the microneedle treatment also induced proliferation of regulatory T cells and a reduction in inflammation.

The researchers designed the microneedle patches so that after releasing their drug payload, they can also collect samples that could be used to monitor the progress of the treatment. Hyaluronic acid causes the needles to swell about tenfold after entering the skin, which allows them to absorb interstitial fluid containing biomolecules and immune cells from the skin.

Following patch removal, researchers can analyze samples to measure levels of regulatory T cells and inflammation markers. This could prove valuable for monitoring future patients who may undergo this treatment.

The researchers now plan to further develop this approach for treating alopecia, and to expand into other autoimmune skin diseases.

The research was funded by the Ignite Fund and Shark Tank Fund awards from the Department of Medicine at Brigham and Women’s Hospital.

Share this news article on:

Related links.

• Natalie Artzi
• Institute for Medical Engineering and Science

Related Topics

• Drug delivery
• Health sciences and technology
• Institute for Medical Engineering and Science (IMES)

Related Articles

A sprayable gel could make minimally invasive surgeries simpler and safer

Patch that delivers drug, gene, and light-based therapy to tumor sites shows promising results

MIT researchers design tailored tissue adhesives

Previous item Next item

More MIT News

Four from MIT named 2024 Knight-Hennessy Scholars

Read full story →

Taking RNAi from interesting science to impactful new treatments

The power of App Inventor: Democratizing possibilities for mobile applications

Using MRI, engineers have found a way to detect light deep in the brain

From steel engineering to ovarian tumor research

A better way to control shape-shifting soft robots

• More news on MIT News homepage →

Massachusetts Institute of Technology 77 Massachusetts Avenue, Cambridge, MA, USA

• Map (opens in new window)
• Events (opens in new window)
• People (opens in new window)
• Careers (opens in new window)
• Accessibility
• Social Media Hub
• MIT on Facebook
• MIT on YouTube
• MIT on Instagram

Novo Nordisk ties up with Metaphore to develop new obesity drugs

• Medium Text

• Company Metaphore Pharmaceuticals, Inc. Follow
• Company Novo Nordisk A/S Follow
• Company Flagship Pioneering, Inc. Follow

WHY IT'S IMPORTANT

By the numbers.

Sign up here.

Reporting by Puyaan Singh in Bengaluru; Editing by Anil D'Silva and Shounak Dasgupta

Our Standards: The Thomson Reuters Trust Principles. New Tab , opens new tab

Business Chevron

OpenAI unveils new AI model as competition heats up

ChatGPT maker OpenAI said on Monday it would release a new AI model called GPT-4o, capable of realistic voice conversation and able to interact across text and image, its latest move to stay ahead in a race to dominate the emerging technology.

• Artificial Intelligence

Google DeepMind’s Latest AI Model Is Poised to Revolutionize Drug Discovery

R esearchers at Google DeepMind have developed AlphaFold 3, an AI model that can predict the structure of and interactions between biological molecules including proteins, DNA and RNA, and small molecules that could function as drugs. Google DeepMind will make the model available for non-commercial use through AlphaFold server . The landmark innovation, the details of which were published in the journal Nature on May 8, is likely to dramatically accelerate biological research. 

“It's a big milestone for us today, announcing AlphaFold 3,” said Demis Hassabis, CEO of Google DeepMind, at a briefing on May 7 announcing the breakthrough. “Biology is a dynamic system and you have to understand how properties of biology emerge through the interactions between different molecules in the cell. You can think of AlphaFold 3 as our first big step towards that.”

The AI system is a descendant of previous AlphaFold models built by Google DeepMind that essentially solved the problem of predicting the three-dimensional structure of a protein from its amino acid structure. Google DeepMind’s first AlphaFold model, announced in 2018, attempted to predict protein structures, coming first in an international protein structure prediction competition. AlphaFold 2, released in 2020, significantly improved on the first’s protein structure accuracy predictions. 

AlphaFold 3 goes further by predicting the structures of almost all biological molecules and modeling the interactions between those molecules. While researchers have long developed specialized computational methods for modeling interactions between specific types of biological molecules, AlphaFold 3 marks the first time that a single system has been able to predict the interactions between nearly all molecular types with state-of-the-art performance.

More From TIME

The properties and functions of molecules in biological systems are typically a result of how they interact with other molecules. Using experiments to understand molecular interactions can take years of research time and be prohibitively expensive. If these interactions can instead be estimated computationally with sufficient accuracy, then biological research can be dramatically accelerated. For example, if researchers believe that a molecule that binds to a specific site on a certain protein would be a promising drug candidate, they can use computational systems such as AlphaFold 3 to test potential drug molecules.

“AlphaFold continues to get better, and increasingly more relevant for biological investigations,” said Paul Nurse, the Nobel Prize-winning geneticist and chief executive and director of the London-based biomedical research center the Francis Crick Institute, in a statement accompanying the Google DeepMind announcement. “This third version will enable increased accuracy in predicting the structures of complexes between different macromolecules, as well as associations between macromolecules, small molecules and ions.”

Google DeepMind was founded as DeepMind in 2010 by Hassabis, along with Google DeepMind Chief AGI Scientist Shane Legg and Mustafa Suleyman . (Suleyman is now CEO at Microsoft AI, Microsoft’s consumer AI products and research organization.) DeepMindwas acquired by Google in 2014, and in 2023 Google merged DeepMind with Google Brain, another Google AI division, to form Google DeepMind, putting an end to efforts by DeepMind’s leadership to secure greater autonomy from their parent company.

In addition to the AlphaFold family of AI systems, Google DeepMind has made several breakthroughs that use AI to further science and technology. In 2022 the company released an AI system that can discover novel algorithms, and in 2023 it released an AI model that could forecast the weather with unprecedented accuracy. Also in 2023, Google DeepMind released an AI model that it claims accurately predicts the structures of materials, although the utility of this model has since been called into question by independent researchers.

In 2021, Google parent company Alphabet announced the creation of Isomorphic Labs, which aims to take an AI-first approach to drug discovery. Researchers from Isomorphic Labs contributed to the development of AlphaFold 3 and, while AlphaFold Server can be used by anyone from non-commercial research, researchers at Isomorphic Labs will have exclusive access to AlphaFold 3 for commercial use.

“We've been using AlphaFold 3’s capabilities day-to-day in our drug design programmes,” said Max Jaderberg, chief AI officer at Isomorphic labs, at the announcement briefing. “We're already seeing that potential to accelerate, improve, and ultimately transform the way that we do drug discovery, and it's really because of the new level of accuracy of this model, and the increased breadth of biomolecules that this model is able to predict, that really enables that for us.”

More Must-Reads From TIME

• What Student Photojournalists Saw at the Campus Protests
• Women Say They Were Pressured Into Long-Term Birth Control
• How Far Trump Would Go
• Scientists Are Finding Out Just How Toxic Your Stuff Is
• Boredom Makes Us Human
• John Mulaney Has What Late Night Needs
• The 100 Most Influential People of 2024
• Want Weekly Recs on What to Watch, Read, and More? Sign Up for Worth Your Time

Write to Will Henshall at [email protected]

AlphaFold 3 predicts the structure and interactions of all of life’s molecules

Introducing AlphaFold 3, a new AI model developed by Isomorphic Labs and Google DeepMind. By accurately predicting the structure of proteins, DNA, RNA, ligands and more, and how they interact, we hope it will help to transform our understanding of the biological world and drug discovery.

Inside every plant, animal, and human cell are billions of molecular machines. They’re made up of proteins, DNA, and other molecules, but no single piece works on its own. Only by seeing how they interact together, across millions of types of combinations, can we start to truly understand life’s processes.

In a paper published in Nature , we introduce AlphaFold 3, a revolutionary model that can predict the structure and interactions of all life’s molecules with unprecedented accuracy. For the interactions of proteins with other molecule types we see at least a 50% improvement compared with existing prediction methods, and for some important categories of interaction we have doubled prediction accuracy.

We hope AlphaFold 3 will help transform our understanding of the biological world and drug discovery. Scientists can access the majority of its capabilities, for free, through the newly launched AlphaFold Server , an easy-to-use research tool. To build on AlphaFold 3’s potential for drug design, we at Isomorphic Labs are already collaborating with pharmaceutical companies to apply it to real-world drug design challenges and, ultimately, develop new life-changing treatments for patients. 

Our new model builds on the foundations of AlphaFold 2, which in 2020 made a fundamental breakthrough in protein structure prediction . So far, millions of researchers globally have used AlphaFold 2 to make discoveries in areas including malaria vaccines, cancer treatments, and enzyme design. AlphaFold has been cited more than 20,000 times and its scientific impact recognized through many prizes, most recently the Breakthrough Prize in Life Sciences . AlphaFold 3 takes us beyond proteins to a broad spectrum of biomolecules. This leap could unlock more transformative science, from accelerating drug design and genomics research, to developing biorenewable materials and more resilient crops.

7PNM - Spike protein of a common cold virus (Coronavirus OC43) : AlphaFold 3’s structural prediction for a spike protein (blue) of a cold virus as it interacts with antibodies (turquoise) and simple sugars (yellow), accurately matches the true structure (gray). The animation shows the protein interacting with an antibody, then a sugar. Advancing our knowledge of such immune-system processes helps better understand coronaviruses, including COVID-19, raising possibilities for improved treatments.

How AlphaFold 3 reveals life’s molecules 

Given an input list of molecules, AlphaFold 3 generates their joint 3D structure, revealing how they all fit together. It models large biomolecules such as proteins, DNA, and RNA, as well as small molecules, also known as ligands - a category encompassing many drugs. Furthermore, AlphaFold 3 can model chemical modifications to these molecules which control the healthy functioning of cells, that when disrupted can lead to disease.

AlphaFold 3’s capabilities come from its next-generation architecture and training that now covers all of life’s molecules. At the core of the model is an improved version of our Evoformer module – a deep learning architecture that underpinned AlphaFold 2’s incredible performance. After processing the inputs, AlphaFold 3 assembles its predictions using a diffusion network, akin to those found in AI image generators. The diffusion process starts with a cloud of atoms, and over many steps converges on its final, most accurate molecular structure.

AlphaFold 3’s predictions of molecular interactions surpass the accuracy of all existing systems. As a single model that computes entire molecular complexes in a holistic way, it’s uniquely able to unify scientific insights.

Read our paper in Nature

7BBV - Enzyme : AlphaFold 3’s prediction for a molecular complex featuring an enzyme protein (blue), an ion (yellow sphere) and simple sugars (yellow), along with the true structure (gray). This enzyme is found in a soil-borne fungus (Verticillium dahliae) that damages a wide range of plants. Insights into how this enzyme interacts with plant cells could help researchers develop healthier, more resilient crops.

Leading drug discovery at Isomorphic Labs

AlphaFold 3 creates capabilities for drug design with predictions for molecules commonly used in drugs, such as ligands and antibodies, that bind to proteins to change how they interact in human health and disease. 

AlphaFold 3 achieves unprecedented accuracy in predicting drug-like interactions, including the binding of proteins with ligands and antibodies with their target proteins. AlphaFold 3 is 50% more accurate than the best traditional methods on the PoseBusters benchmark , without needing the input of any structural information, making AlphaFold 3 the first AI system to surpass physics-based tools for biomolecular structure prediction. The ability to predict antibody-protein binding is critical to understanding aspects of the human immune response and the design of new antibodies - a growing class of therapeutics.

Using AlphaFold 3 in combination with a complementary suite of in-house AI models, we are working on drug design for internal projects as well as with pharmaceutical partners. We are using AlphaFold 3 to accelerate and improve the success of drug design - by helping understand how to approach new disease targets, and developing novel ways to pursue existing ones that were previously out of reach. ‍

Read more about how we are using AlphaFold 3 for drug design.

AlphaFold Server: A free and easy-to-use research tool 

8AW3 - RNA modifying protein : AlphaFold 3’s prediction for a molecular complex featuring a protein (blue), a strand of RNA (purple), and two ions (yellow) closely matches the true structure (gray). This complex is involved with the creation of other proteins - a cellular process fundamental to life and health.

Google DeepMind’s newly launched AlphaFold Server is the most accurate tool in the world for predicting how proteins interact with other molecules throughout the cell. It is a free platform that scientists around the world can use for non-commercial research. With just a few clicks, biologists can harness the power of AlphaFold 3 to model structures composed of proteins, DNA, RNA, and a selection of ligands, ions, and chemical modifications.

AlphaFold Server helps scientists make novel hypotheses to test in the lab, speeding up workflows and enabling further innovation. This gives researchers an accessible way to generate predictions, regardless of their access to computational resources or their expertise in machine learning.  

Experimental protein-structure prediction can take about the length of a PhD and cost hundreds of thousands of dollars. Google DeepMind's previous model, AlphaFold 2, has been used to predict hundreds of millions of structures, which would have taken hundreds of millions of researcher-years at the current rate of experimental structural biology.

"With AlphaFold Server, it’s not only about predicting structures anymore, it’s about generously giving access: allowing researchers to ask daring questions and accelerate discoveries.”

Céline Bouchoux, The Francis Crick Institute

Explore AlphaFold Server

Sharing the power of AlphaFold 3 responsibly

Alongside Google DeepMind, we’ve sought to understand the broad impact of the technology. Working together with the research and safety community to take a science-led approach, we have conducted extensive assessments to mitigate potential risks and share the widespread benefits to biology and humanity.

Building on the external consultations we carried out for AlphaFold 2, Google DeepMind have now engaged with more than 50 domain experts, in addition to specialist third parties, across biosecurity, research, and industry, to understand the capabilities of successive AlphaFold models and any potential risks. We also participated in community-wide forums and discussions ahead of AlphaFold 3’s launch.

AlphaFold Server reflects the ongoing commitment to share the benefits of AlphaFold, including the free database of 200 million protein structures. We’ll continue to work with the scientific community and policy makers to develop and deploy AI technologies responsibly.

Opening up the future of AI-powered cell biology

7R6R - DNA binding protein : AlphaFold 3’s prediction for a molecular complex featuring a protein (blue) bound to a double helix of DNA (pink) is a near-perfect match to the true molecular structure discovered through painstaking experiments (gray).

AlphaFold 3 brings the biological world into high definition. It allows scientists to see cellular systems in all their complexity, across structures, interactions, and modifications. This new window on the molecules of life reveals how they’re all connected and helps understand how those connections affect biological functions – such as the actions of drugs, the production of hormones, and the health-preserving process of DNA repair. 

The impacts of AlphaFold 3 and the free AlphaFold Server will be realised through how they empower scientists to accelerate discovery across open questions in biology and new lines of research. We’re just beginning to tap into AlphaFold 3’s potential and can’t wait to see what the future holds.

Learn more: 

Read our blog on Rational Drug Design with AlphaFold 3

Read the Isomorphic Labs blog

Latest from iso.