Skip to content
DraconiticResearch
Search Sources API
203,234 studies31 sourcesUpdated 46 min ago
Peer-reviewedCC BY-NC-NDNature Biotechnology

AI-enhanced adaptive virtual screening of large libraries for ligand discovery

  • Domiziana Cecchini1
  • AkshatKumar Nigam2
  • Ming Tang3,4,5
  • Joana Reis6
  • Matt Koop7
  • Andrea Gottinger1
  • Callum R. Nicoll1
  • Yao Wang8
  • Abhilash Jayaraj9,6
  • Süleyman Selim Çınaroğlu10,11
  • Ricarda Törner6,12
  • Yehor S. Malets13
36 more authors
  • Minko Gehev14
  • Krishna M. Padmanabha Das9,3
  • Kelly A. Churion3
  • Jongwan Kim3
  • Nidhin Thomas3
  • Yong Li15
  • Hyuk‐Soo Seo16
  • Sirano Dhe‐Paganon16
  • Christopher Secker17,18
  • Mohammad Haddadnia9,6,19,20,21,22
  • Alexander Hasson23,24
  • Minkai Li25
  • Abhishek Kumar26,3,27
  • Roni Levin‐Konigsberg2,28
  • Eun‐Bee Choi29
  • Geoffrey Ira Shapiro29
  • Huel Cox6
  • Luke Sebastian6
  • Chelsea Braithwaite6
  • Puspalata Bashyal6
  • Dmytro S. Radchenko30
  • Aditya Kumar26,3,27
  • Lei Yang15
  • Pierre-Yves Aquilanti7,31
  • Henry A. Gabb32
  • Amr Alhossary33,34
  • Eric O’Neill23
  • Gerhard Wagner9
  • Alán Aspuru‐Guzik35,36,37,19,20,38,39,40,22
  • Yurii S. Moroz30
  • Charalampos G. Kalodimos3
  • Konstantin Fackeldey27,18
  • John D. Schuetz8
  • Andrea Mattevi
  • Haribabu Arthanari
  • Christoph Gorgulla
40 affiliations
  1. 1Department of Biology and Biotechnology, University of Pavia, Pavia, Italy
  2. 2Department of Computer Science, Stanford University, Stanford, CA, USA
  3. 3Department of Structural Biology, St. Jude Children's Research Hospital, Memphis, TN, USA
  4. 4Frazer Institute, Faculty of Medicine at the Translational Research Institute Australia, The University of Queensland, Brisbane, Queensland, Australia
  5. 5School of Biomedical Sciences, Centre for Genomics and Personalised Health at the Translational Research Institute Australia, Queensland University of Technology, Brisbane, Queensland, Australia
  6. 6Department of Cancer Biology, Dana-Farber Cancer Institute, Boston, MA, USA
  7. 7Amazon Web Services, Seattle, WA, USA
  8. 8Department of Pharmacy and Pharmaceutical Sciences, St. Jude Children's Research Hospital, Memphis, TN, USA
  9. 9Department of Biological Chemistry and Molecular Pharmacology, Harvard Medical School, Harvard University, Boston, MA, USA
  10. 10Department of Biochemistry, University of Oxford, Oxford, UK
  11. 11Department of Bioengineering, Ege University, Bornova, Türkiye
  12. 12Department of Chemistry, University of Zurich, Zurich, Switzerland
  13. 13V. P. Kukhar Institute of Bioorganic Chemistry and Petrochemistry, National Academy of Science of Ukraine, Kyiv, Ukraine
  14. 14Google, Mountain View, CA, USA
  15. 15Department of Chemical Biology and Therapeutics, St. Jude Children's Research Hospital, Memphis, TN, USA
  16. 16Chemical Biology Program, Dana-Farber Cancer Institute, Boston, MA, USA
  17. 17Max Delbrück Center for Molecular Medicine, Berlin, Germany
  18. 18Zuse Institute Berlin (ZIB), Berlin, Germany
  19. 19Department of Chemistry, University of Toronto, Toronto, Ontario, Canada
  20. 20Department of Computer Science, University of Toronto, Toronto, Ontario, Canada
  21. 21Department of Molecular Genetics, University of Toronto, Toronto, Ontario, Canada
  22. 22Vector Institute for Artificial Intelligence, Toronto, Ontario, Canada
  23. 23Department of Oncology, University of Oxford, Oxford, UK
  24. 24Mathematical Institute, University of Oxford, Oxford, UK
  25. 25Harvard College, Cambridge, MA, USA
  26. 26Department of Biosciences and Bioengineering, Indian Institute of Technology Guwahati, Guwahati, India
  27. 27Institute for Mathematics, Technical University Berlin, Berlin, Germany
  28. 28Department of Genetics, Stanford University, Stanford, CA, USA
  29. 29Department of Medical Oncology, Dana-Farber Cancer Institute, Boston, MA, USA
  30. 30Enamine Ltd, Kyiv, Ukraine
  31. 31NVIDIA, Santa Clara, CA, USA
  32. 32Intel, Santa Clara, CA, USA
  33. 33Pulse for Integrated Solutions, Unterschleißheim, Germany
  34. 34Wesleyan University, Middletown, CT, USA
  35. 35Acceleration Consortium, Acceleration Consortium, Toronto, Ontario, Canada
  36. 36Canadian Institute for Advanced Research, Toronto, Ontario, Canada
  37. 37Department of Chemical Engineering & Applied Chemistry, University of Toronto, Toronto, Ontario, Canada
  38. 38Department of Materials Science & Engineering, University of Toronto, Toronto, Ontario, Canada
  39. 39Institute of Medical Science, University of Toronto, Toronto, Ontario, Canada
  40. 40NVIDIA, Toronto, Ontario, Canada
Open at sourcePDFDOICite

Abstract

Identifying potent lead molecules for specific targets remains a major bottleneck in drug discovery. As structural information about proteins becomes increasingly available, ultra-large virtual screenings (ULVSs) which computationally evaluate billions of molecules offer a powerful way to accelerate early-stage drug discovery. Here, we introduce AdaptiveFlow, an open-source platform designed to make ULVSs more accessible, scalable, and efficient. AdaptiveFlow provides free access to a screening-ready version of the Enamine REAL Space, the largest library of ready-to-dock, drug-like molecules, containing 69 billion compounds that we prepared using the ligand preparation module of the platform. A key innovation of the platform is its use of a multi-dimensional grid of molecular properties, which helps researchers explore and prioritize chemical space more effectively and reduce the computational costs by a factor of approximately 1000. This grid forms the basis of a new method for identifying promising regions of chemical space, enabling systematic exploration and prioritization of compound libraries. An optional active learning component can further accelerate this process by adaptively steering the search toward molecules most likely to bind a given target. To support a broad range of applications, AdaptiveFlow is compatible with over 1,500 docking methods. The platform achieves near-linear scaling on up to 5.6 million CPUs in the AWS Cloud, setting a new benchmark for large-scale cloud computing in drug discovery. Using this approach, we identified nanomolar inhibitors of two disease-relevant targets: ferroptosis suppressor protein 1 (FSP1) and poly(ADP-ribose) polymerase 1 (PARP-1). By leveraging newly solved crystal structures of FSP1 in complex with NAD+, FAD, and coenzyme Q1, we validated these hits experimentally and determined the first co-crystal structures of FSP1 bound to small-molecule inhibitors, enabling insights into inhibitor binding mechanisms previously unknown. With its high scalability, flexibility, and open accessibility, AdaptiveFlow offers a powerful new resource for discovering and optimizing drug candidates at an unprecedented scale and speed.

Related

  1. A Pharmacophore-Based Virtual Screening Approach for the Discovery of Novel P2Y ₁₂ Receptor Ligands
    Peter H. Moons, Anne van der Schatte van der Schatte Olivier, Fabian Dey +7
    ChemRxivPreprintLicence not statedGeneral chemistryChemRxiv
  2. A New Perspective in Chemical Probe Discovery: Expanding Virtual Screening into Metal Complex Chemical Space
    Mika Kintzel, Yew Mun Yip, Daniella Hares +12
    ChemRxivPreprintLicence not statedGeneral chemistryChemRxiv
  3. On-resin Heck coupling enables small-molecule ligand discovery with covalent peptide-encoded libraries
    Ching-Pei Hsu, Michael Desgagné, Simon L. Rössler +8
    Chemical SciencePeer-reviewedCC BY-NC 3.0Full textOrganic chemistryPMC
  4. DEL-GPT: learning the language of DNA-encoded libraries to design focused screening collections
    Akhila Mettu, Naveed Naemi, Raphael Franzini +1
    ChemRxivPreprintCC BYFull textGeneral chemistryChemRxiv
  5. The Virtual Biotech: A multi-agent AI framework for therapeutic discovery and development
    Harrison G. Zhang, Peter Eckmann, Jiacheng Miao +2
    SciencePeer-reviewedLicence not stated16 citedMedicinal chemistry & drug discoveryOpenAlex

Cite

BibTeXRIS
BibTeX
@article{cecchini2026aienhanced,
  title = {{AI-enhanced adaptive virtual screening of large libraries for ligand discovery}},
  author = {Domiziana Cecchini and AkshatKumar Nigam and Ming Tang and Joana Reis and Matt Koop and Andrea Gottinger and Callum R. Nicoll and Yao Wang and Abhilash Jayaraj and Süleyman Selim Çınaroğlu and Ricarda Törner and Yehor S. Malets and Minko Gehev and Krishna M. Padmanabha Das and Kelly A. Churion and Jongwan Kim and Nidhin Thomas and Yong Li and Hyuk‐Soo Seo and Sirano Dhe‐Paganon and Christopher Secker and Mohammad Haddadnia and Alexander Hasson and Minkai Li and Abhishek Kumar and Roni Levin‐Konigsberg and Eun‐Bee Choi and Geoffrey Ira Shapiro and Huel Cox and Luke Sebastian and Chelsea Braithwaite and Puspalata Bashyal and Dmytro S. Radchenko and Aditya Kumar and Lei Yang and Pierre-Yves Aquilanti and Henry A. Gabb and Amr Alhossary and Eric O’Neill and Gerhard Wagner and Alán Aspuru‐Guzik and Yurii S. Moroz and Charalampos G. Kalodimos and Konstantin Fackeldey and John D. Schuetz and Andrea Mattevi and Haribabu Arthanari and Christoph Gorgulla},
  journal = {Nature Biotechnology},
  year = {2026},
  month = sep,
  doi = {10.1038/s41587-026-03217-x},
  url = {https://doi.org/10.1038/s41587-026-03217-x},
  note = {CC BY-NC-ND},
}
RIS
TY  - JOUR
TI  - AI-enhanced adaptive virtual screening of large libraries for ligand discovery
AU  - Domiziana Cecchini
AU  - AkshatKumar Nigam
AU  - Ming Tang
AU  - Joana Reis
AU  - Matt Koop
AU  - Andrea Gottinger
AU  - Callum R. Nicoll
AU  - Yao Wang
AU  - Abhilash Jayaraj
AU  - Süleyman Selim Çınaroğlu
AU  - Ricarda Törner
AU  - Yehor S. Malets
AU  - Minko Gehev
AU  - Krishna M. Padmanabha Das
AU  - Kelly A. Churion
AU  - Jongwan Kim
AU  - Nidhin Thomas
AU  - Yong Li
AU  - Hyuk‐Soo Seo
AU  - Sirano Dhe‐Paganon
AU  - Christopher Secker
AU  - Mohammad Haddadnia
AU  - Alexander Hasson
AU  - Minkai Li
AU  - Abhishek Kumar
AU  - Roni Levin‐Konigsberg
AU  - Eun‐Bee Choi
AU  - Geoffrey Ira Shapiro
AU  - Huel Cox
AU  - Luke Sebastian
AU  - Chelsea Braithwaite
AU  - Puspalata Bashyal
AU  - Dmytro S. Radchenko
AU  - Aditya Kumar
AU  - Lei Yang
AU  - Pierre-Yves Aquilanti
AU  - Henry A. Gabb
AU  - Amr Alhossary
AU  - Eric O’Neill
AU  - Gerhard Wagner
AU  - Alán Aspuru‐Guzik
AU  - Yurii S. Moroz
AU  - Charalampos G. Kalodimos
AU  - Konstantin Fackeldey
AU  - John D. Schuetz
AU  - Andrea Mattevi
AU  - Haribabu Arthanari
AU  - Christoph Gorgulla
PY  - 2026
DA  - 2026/09/01
JO  - Nature Biotechnology
DO  - 10.1038/s41587-026-03217-x
UR  - https://doi.org/10.1038/s41587-026-03217-x
AB  - Identifying potent lead molecules for specific targets remains a major bottleneck in drug discovery. As structural information about proteins becomes increasingly available, ultra-large virtual screenings (ULVSs) which computationally evaluate billions of molecules offer a powerful way to accelerate early-stage drug discovery. Here, we introduce AdaptiveFlow, an open-source platform designed to make ULVSs more accessible, scalable, and efficient. AdaptiveFlow provides free access to a screening-ready version of the Enamine REAL Space, the largest library of ready-to-dock, drug-like molecules, containing 69 billion compounds that we prepared using the ligand preparation module of the platform. A key innovation of the platform is its use of a multi-dimensional grid of molecular properties, which helps researchers explore and prioritize chemical space more effectively and reduce the computational costs by a factor of approximately 1000. This grid forms the basis of a new method for identifying promising regions of chemical space, enabling systematic exploration and prioritization of compound libraries. An optional active learning component can further accelerate this process by adaptively steering the search toward molecules most likely to bind a given target. To support a broad range of applications, AdaptiveFlow is compatible with over 1,500 docking methods. The platform achieves near-linear scaling on up to 5.6 million CPUs in the AWS Cloud, setting a new benchmark for large-scale cloud computing in drug discovery. Using this approach, we identified nanomolar inhibitors of two disease-relevant targets: ferroptosis suppressor protein 1 (FSP1) and poly(ADP-ribose) polymerase 1 (PARP-1). By leveraging newly solved crystal structures of FSP1 in complex with NAD+, FAD, and coenzyme Q1, we validated these hits experimentally and determined the first co-crystal structures of FSP1 bound to small-molecule inhibitors, enabling insights into inhibitor binding mechanisms previously unknown. With its high scalability, flexibility, and open accessibility, AdaptiveFlow offers a powerful new resource for discovering and optimizing drug candidates at an unprecedented scale and speed.
SN  - 0733-222X
SN  - 1087-0156
SN  - 1546-1696
SN  - 2331-3684
ER  -