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

Ultralarge virtual screenings (ULVSs) evaluate billions of molecules for drug discovery but face cost, flexibility and scalability limits. We introduce AdaptiveFlow, an open-source platform that makes ULVSs more accessible, scalable and efficient and supports artificial intelligence (AI) and machine learning (ML) method development. AdaptiveFlow provides a screening-ready version of the Enamine REAL Space, to our knowledge the largest library of ready-to-dock, drug-like molecules, comprising 69 billion compounds, also available in SELFIES format. An 18-dimensional grid of molecular properties prioritizes promising chemical subspaces, with optional active learning, reducing computational costs by orders of magnitude. AdaptiveFlow integrates >1,500 docking protocols, including GPU-accelerated and ML-based methods, and achieves near-linear scaling on up to 5.6 million CPUs in the Amazon Web Services cloud. We identified nanomolar inhibitors of two disease-relevant targets, ferroptosis suppressor protein 1 (FSP1) and poly(ADP-ribose) polymerase 1. Co-crystal structures provided mechanistic insights into FSP1 inhibition. AdaptiveFlow enables drug discovery at unprecedented scale and supports the development of AI-driven methods.

Authors

Institutions

Publication Details

Journal
Nature Biotechnology
Published
2026-09-01
DOI
https://doi.org/10.1038/s41587-026-03217-x
Primary Topic
Machine Learning in Materials Science
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
article

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

Süleyman Selim Çınaroğlu, A. Kumar, Yurii S. Moroz, Roni Levin‐Konigsberg et al.
Nature Biotechnology
Machine Learning in Materials Science
article

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

Süleyman Selim Çınaroğlu, A. Kumar, Yurii S. Moroz, Roni Levin‐Konigsberg, Alexander Hasson, Andrea Mattevi, Lei Yang, Puspalata Bashyal, AkshatKumar Nigam, Haribabu Arthanari, Huel Cox, Pierre-Yves Aquilanti, Eun‐Bee Choi, Christoph Gorgulla, Sirano Dhe‐Paganon, Konstantin Fackeldey, Mohammad Haddadnia, Henry A. Gabb, Geoffrey I. Shapiro, Abhilash Jayaraj, Andrea Gottinger, Callum R. Nicoll, Dmytro S. Radchenko, John D. Schuetz, Domiziana Cecchini, Nidhin Thomas, Matt Koop, Alán Aspuru‐Guzik, Hyuk‐Soo Seo, Luke Sebastian, Christopher Secker, Yehor S. Malets, Amr Alhossary, Ricarda Törner, Ming Tang, Gerhard Wagner, Yong Li, Jongwan Kim, Aditya Kumar, Chelsea Braithwaite, Joana Reis, Kelly Churion, Krishna M. Padmanabha Das, Minko Gehev, Minkai Li, Yao Wang, Charalampos G. Kalodimos, Eric O’Neill
article en

Abstract

Ultralarge virtual screenings (ULVSs) evaluate billions of molecules for drug discovery but face cost, flexibility and scalability limits. We introduce AdaptiveFlow, an open-source platform that makes ULVSs more accessible, scalable and efficient and supports artificial intelligence (AI) and machine learning (ML) method development. AdaptiveFlow provides a screening-ready version of the Enamine REAL Space, to our knowledge the largest library of ready-to-dock, drug-like molecules, comprising 69 billion compounds, also available in SELFIES format. An 18-dimensional grid of molecular properties prioritizes promising chemical subspaces, with optional active learning, reducing computational costs by orders of magnitude. AdaptiveFlow integrates >1,500 docking protocols, including GPU-accelerated and ML-based methods, and achieves near-linear scaling on up to 5.6 million CPUs in the Amazon Web Services cloud. We identified nanomolar inhibitors of two disease-relevant targets, ferroptosis suppressor protein 1 (FSP1) and poly(ADP-ribose) polymerase 1. Co-crystal structures provided mechanistic insights into FSP1 inhibition. AdaptiveFlow enables drug discovery at unprecedented scale and supports the development of AI-driven methods.

Nature Biotechnology
Wesleyan University (US), Broad Institute (US), Canadian Institute for Advanced Research (CA), Google (United States) (US), Nvidia (United Kingdom) (GB), Translational Research Institute (AU), Amazon (United States) (US), St. Jude Children's Research Hospital (US), Indian Institute of Technology Guwahati (IN), Intel (United States) (US), Harvard University (US), Queensland University of Technology (AU), The University of Queensland (AU), University of Toronto (CA), Taras Shevchenko National University of Kyiv (UA), Zuse Institute Berlin (DE), University of Zurich (CH), Max Delbrück Center (DE), University of Pavia (IT), Harvard University Press (US), Structural Genomics Consortium (CA), University of Oxford (GB), Ege University (TR), Harvard College Observatory (US), Institute of Bioorganic Chemistry and Petrochemistry V.P. Kukhar (UA), Dana-Farber Cancer Institute (US), Integrated Solutions for Systems (United States) (US), Vector Institute (CA), Nvidia (United States) (US), Dana-Farber/Harvard Cancer Center (US), Enamine (Ukraine) (UA), Stanford University (US)
St. Jude Children's Research Hospital, American Lebanese Syrian Associated Charities, Deutsche Forschungsgemeinschaft, Associazione Italiana per la Ricerca sul Cancro, National Institutes of Health, National Institute of General Medical Sciences
Openalex Percentile: Top 24%
Machine Learning in Materials Science
DOIOpen Access PDFOpenAlex
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.