pocketz.doi.bio/2026-05-15_182553_linkspringercom_978-3-032-04552-2_5

Uni-Mol Docking V2: Towards Realistic and Accurate Binding Pose Prediction | Springer Nature Link

URL: https://link.springer.com/chapter/10.1007/978-3-032-04552-2_5

Saved: 2026-05-15T01:25:58.663Z

Eric Alcaide, Zhifeng Gao, Guolin Ke, Yaqi Li, Linfeng Zhang, Hang Zheng & Gengmo Zhou

Part of the book series: Lecture Notes in Computer Science ((LNCS,volume 16072))

Included in the following conference series:

International Conference on Artificial Neural Networks

605 Accesses

15 Citations

Abstract

In recent years, machine learning (ML) methods have emerged as promising alternatives for molecular docking, offering the potential for high accuracy without incurring prohibitive computational costs. However, recent studies have indicated that these ML models may overfit to quantitative metrics while neglecting the physical constraints inherent in the problem. In this work, we present Uni-Mol Docking V2, which demonstrates a remarkable improvement in performance, accurately predicting the binding poses of 77+% of ligands in the PoseBusters benchmark with an RMSD value of less than 2.0 Å, and 75+% passing all quality checks. This represents a significant increase from the 62% achieved by the previous Uni-Mol Docking model. Notably, our Uni-Mol Docking approach generates chemically accurate predictions, circumventing issues such as chirality inversions and steric clashes that have plagued previous ML models. Furthermore, we observe enhanced performance in terms of high-quality predictions (RMSD values of less than 1.0 Åand 1.5 Å) and physical soundness when Uni-Mol Docking is combined with more physics-based methods like Uni-Dock. Our results represent a significant advancement in the application of artificial intelligence for scientific research, adopting a holistic approach to ligand docking that is well-suited for industrial applications in virtual screening and drug design. The code, data and service for Uni-Mol Docking are publicly available for use and further development in https://github.com/dptech-corp/Uni-Mol.

This is a preview of subscription content, log in via an institution to check access.

Access this chapter Log in via an institution Subscribe and save Springer+ from $39.99 /Month Starting from 10 chapters or articles per month Access and download chapters and articles from more than 300k books and 2,500 journals Cancel anytime View plans Buy Now Chapter USD 29.95 Price excludes VAT (Canada) Available as PDF Read on any device Instant download Own it forever Buy Chapter eBook USD 59.99 Price excludes VAT (Canada) Available as EPUB and PDF Read on any device Instant download Own it forever Buy eBook Softcover Book USD 79.99 Price excludes VAT (Canada) Compact, lightweight edition Dispatched in 3 to 5 business days Free shipping worldwide - see info Buy Softcover Book

Tax calculation will be finalised at checkout

Purchases are for personal use only

Institutional subscriptions

Similar content being viewed by others Predicting target profiles with confidence as a service using docking scores Article Open access 15 October 2020 MC-GNNAS-Dock: Multi-criteria GNN-Based Algorithm Selection for Molecular Docking Chapter © 2026 Effect of preprocessing and simulation parameters on the performance of molecular docking studies Article 15 July 2023 Explore related subjects Discover the latest articles, books and news in related subjects, suggested using machine learning. Computational Chemistry Molecular Modelling Protein structure predictions Protein-Ligand Interactions Structure Prediction Supramolecular Assembly Protein Structure Prediction and Function Analysis References

Alcaide, E., Li, Z., Zheng, H., Gao, Z., Ke, G.: UMD-fit: generating realistic ligand conformations for distance-based deep docking models. In: NeurIPS 2023 Generative AI and Biology (GenBio) Workshop (2023). https://openreview.net/forum?id=BE2hok0lES

Bryant, P., Kelkar, A., Guljas, A., Clementi, C., Noé, F.: Structure prediction of protein-ligand complexes from sequence information with Umol. Nat. Commun. 15(1), 4536 (2024). https://doi.org/10.1038/s41467-024-48837-6

Buttenschoen, M., Morris, G.M., Deane, C.M.: Posebusters: AI-based docking methods fail to generate physically valid poses or generalise to novel sequences. Chem. Sci. 15, 3130–3139 (2024). https://doi.org/10.1039/D3SC04185A

Article Google Scholar

Hu, L., Benson, M.L., Smith, R.D., Lerner, M.G., Carlson, H.A.: Binding moad (mother of all databases). Proteins: Struct., Funct., Bioinfor. 60(3), 333–340 (2005). https://doi.org/10.1002/prot.20512

Krishna, R., et al.: Generalized biomolecular modeling and design with rosettafold all-atom. Science 384(6693), eadl2528 (2024). https://doi.org/10.1126/science.adl2528, https://www.science.org/doi/abs/10.1126/science.adl2528

Lu, S., Gao, Z., He, D., Zhang, L., Ke, G.: Data-driven quantum chemical property prediction leveraging 3D conformations with Uni-Mol+. Nat. Commun. 15(1), 7104 (2024). https://doi.org/10.1038/s41467-024-51321-w

Article Google Scholar

Plainer, M., et al.: DiffDock-pocket: diffusion for pocket-level docking with sidechain flexibility (2023). https://neurips.cc/virtual/2023/77419

Su, M., et al.: Comparative assessment of scoring functions: the casf-2016 update. J. Chem. Inf. Model. 59(2), 895–913 (2018). https://doi.org/10.1021/acs.jcim.8b00545

Article Google Scholar

Team, I.L., Team, G.D.A.: A glimpse of the next generation of alphafold. https://storage.googleapis.com/deepmind-media/DeepMind.com/Blog/a-glimpse-of-the-next-generation-of-alphafold/alphafoldlatestoct2023.pdf

Trott, O., Olson, A.J.: Autodock vina: improving the speed and accuracy of docking with a new scoring function, efficient optimization, and multithreading. J. Comput. Chem. 31(2), 455–461 (2010). https://doi.org/10.1002/jcc.21334

Article Google Scholar

Yang, H., et al.: Synergistic application of molecular docking and machine learning for improved protein-ligand binding pose prediction. Natl. Sci. Open 3(2), 20230058 (2024). https://doi.org/10.1360/nso/20230058

Article Google Scholar

Zhou, G., et al.: Uni-Mol: a universal 3D molecular representation learning framework. In: The Eleventh International Conference on Learning Representations (2023). https://openreview.net/forum?id=6K2RM6wVqKu

Download references

Acknowledgments

Our work is made possible by the dedication and efforts of numerous teams at DP Technology and AI for Science Institute. We would like to acknowledge the support from MLOps Team, PR Team, Drug Discovery Team etc. Contributors (in alphabetical orders).

Eric Alcaide: Conceptualization, Methodology, Software, Validation, Writing.
Zhifeng Gao: Supervision, Conceptualization, Methodology, Writing.
Guolin Ke: Supervision, Project administration, Conceptualization, Methodology.
Yaqi Li: Data Curation, Investigation, Validation.
Linfeng Zhang: Supervision, Funding acquisition.
Hang Zheng: Software, Methodology, Data Curation, Validation.
Gengmo Zhou: Conceptualization, Methodology, Software, Validation, Writing.

This study was also partially funded by the Horizon Europe funding programme under the Marie Skłodowska-Curie Actions Doctoral Networks grant agreement “Explainable AI for Molecules - AiChemist”, no. 101120466 (to E.A.) and the Swiss State Secretariat for Education, Research and lnnovation (SERI) under contract number 23.00421.

Author information Authors and Affiliations

DP Technology Beijing, Haidian District, Beijing, China

Eric Alcaide, Zhifeng Gao, Guolin Ke, Yaqi Li, Linfeng Zhang, Hang Zheng & Gengmo Zhou

AI for Science Institute Beijing, Haidian District, Beijing, China

Zhifeng Gao, Guolin Ke & Linfeng Zhang

USI, IDSIA, Lugano, Switzerland

Eric Alcaide

Corresponding author

Correspondence to Eric Alcaide .

Editor information Editors and Affiliations

University of Bern, Bern, Switzerland

Walter Senn

University of Genoa, Genoa, Italy

Marcello Sanguineti

Vytautas Magnus University, Kaunas, Lithuania

Ausra Saudargiene

Helmholtz Zentrum München, Neuherberg, Germany

Igor V. Tetko

University of Lausanne, Lausanne, Switzerland

Alessandro E. P. Villa

Aix-Marseille Université, Marseille, France

Viktor Jirsa

Université de Montréal, Montréal, QC, Canada

Yoshua Bengio

Rights and permissions

Reprints and permissions

About this paper Cite this paper

Alcaide, E. et al. (2026). Uni-Mol Docking V2: Towards Realistic and Accurate Binding Pose Prediction. In: Senn, W., et al. Artificial Neural Networks and Machine Learning. ICANN 2025 International Workshops and Special Sessions. ICANN 2025. Lecture Notes in Computer Science, vol 16072. Springer, Cham. https://doi.org/10.1007/978-3-032-04552-2_5

Download citation .RIS.ENW.BIB

DOI https://doi.org/10.1007/978-3-032-04552-2_5

Published 23 September 2025

Publisher Name Springer, Cham

Print ISBN 978-3-032-04551-5

Online ISBN 978-3-032-04552-2

eBook Packages Computer Science Computer Science (R0) Springer Nature Proceedings Computer Science

Publish with us

Policies and ethics