Sampling and Ranking of Protein Conformations Using Machine Learning Techniques Do Not Improve the Quality of Rigid Protein–Protein Docking
Open accessPublished version of a preprint
Abstract
Rigid docking remains the most popular method of predicting protein-protein interactions in cases when experimental 3D structures of the complexes are not available. The docking often relies on known unbound (Apo) protein structures, which may differ significantly from their bound (Holo) forms. Modern machine learning (ML) based conformational sampling techniques allow generating ensembles of functionally relevant protein structures, which may be closer to their Holo forms and thus could improve the outcomes of the classical rigid protein-protein docking. Here, we sampled conformations of the protein subunits in 30 complexes from the novel PINDER data set with two state-of-the-art ML-based techniques and evaluated their docking performance using several physics-based, data-based, and ML-based scoring functions. We showed that such conformational sampling rarely produces structures that are closer to the Holo conformations than the corresponding Apo ones. Moreover, even when such conformations are generated, none of the tested scoring functions were able to prioritize and rank them correctly. Our work highlights critical limitations in the current ML-enhanced rigid protein-protein docking workflows and emphasizes the need for new approaches that can better utilize the potential of modern techniques for conformational generation and scoring.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
Highly accurate protein structure prediction with AlphaFold
Jumper, J., Evans, R., Pritzel, A., Green, T., Figurnov, M., Ronneberger, O., Tunyasuvunakool, K., Bates, R., Žídek, A., Potapenko, A., et al. (2021) · Nature · doi:10.1038/s41586-021-03819-2 · Cited by 43,764 · see citing works
Lu, H., Zhou, Q., He, J., Jiang, Z., Peng, C., Tong, R., & Shi, J. (2020) · Signal Transduction and Targeted Therapy · doi:10.1038/s41392-020-00315-3 · Cited by 891 · see citing works
Csermely, P., Palotai, R., & Nussinov, R. (2010) · Trends in Biochemical Sciences · doi:10.1016/j.tibs.2010.04.009 · Cited by 821 · see citing works
CAPRI: A Critical Assessment of PRedicted Interactions
Janin, J., Henrick, K., Moult, J., Eyck, L. T., Sternberg, M. J. E., Vajda, S., Vakser, I., & Wodak, S. J. (2003) · Proteins: Structure, Function, and Bioinformatics · doi:10.1002/prot.10381 · Cited by 703 · see citing works
Predicting multiple conformations via sequence clustering and AlphaFold2
Wayment-Steele, H. K., Ojoawo, A., Otten, R., Apitz, J. M., Pitsawong, W., Hömberger, M., Ovchinnikov, S., Colwell, L., & Kern, D. (2024) · Nature · doi:10.1038/s41586-023-06832-9 · Cited by 545 · see citing works
Sampling alternative conformational states of transporters and receptors with AlphaFold2
del Alamo, D., Sala, D., Mchaourab, H. S., & Meiler, J. (2022) · eLife · doi:10.7554/elife.75751 · Cited by 511 · see citing works
Scalable emulation of protein equilibrium ensembles with generative deep learning
Lewis, S., Hempel, T., Jiménez-Luna, J., Gastegger, M., Xie, Y., Foong, A. Y. K., Satorras, V. G., Abdin, O., Veeling, B. S., Zaporozhets, I., et al. (2025) · Science · doi:10.1126/science.adv9817 · Cited by 335 · see citing works
High-throughput prediction of protein conformational distributions with subsampled AlphaFold2
Monteiro da Silva, G., Cui, J. Y., Dalgarno, D. C., Lisi, G. P., & Rubenstein, B. M. (2024) · Nature Communications · doi:10.1038/s41467-024-46715-9 · Cited by 196 · see citing works
Blind prediction of homo- and hetero-protein complexes: the CASP13-CAPRI experiment
Lensink, M. F., Brysbaert, G., Nadzirin, N., Velankar, S., Chaleil, R. A. G., Gerguri, T., Bates, P. A., Laine, E., Carbone, A., Grudinin, S., et al. (2019) · Proteins Structure Function and Bioinformatics · doi:10.1002/prot.25838 · Cited by 122 · see citing works
Flagged first, then by citations
- Authors
- Stratiichuk, R., Kyrylenko, R., Koleiev, I., Savchenko, I., Voitsitskyi, T., Husak, V., Yesylevskyy, S., Starosyla, S., & Nafiiev, A.
- Affiliations
- Taras Shevchenko National University of Kyiv (UA)
- Receptor.AI Inc., 20-22 Wenlock Road, London N1 7GU, U.K.
- National Academy of Sciences of Ukraine (UA)
- University of Trento (IT)
- Corresponding author
- Kyrylenko, Roman
- ORCID
- Venue
- Journal of Chemical Information and Modeling
- Publisher
- American Chemical Society (ACS)
- Published
- 2025-09-16
- Volume
- 65
- Issue
- 19
- Pages
- 10167-10179
- ISSN
- 1549-9596 (ISSN-L)
- 1549-960X
- OpenAlex
- W4414250097
- Cited by
- 6
- References
- 9
- Citations by year
- 2026: 5 · 2025: 1
- FWCI
- 1.80
- Type
- journal article
- Language
- en
- Open access
- green
- doi.org
- Licence
- Locations
- Journal of Chemical Information and Modeling (publishedVersion)
- bioRxiv (Cold Spring Harbor Laboratory) (acceptedVersion, cc-by)
- PubMed (publishedVersion)
- Funders
- Ministerstvo ?kolstv?, Ml?de?e a Telov?chovy (355/2025-16)
- H2020 Marie Sklodowska-Curie Actions (101101923)
- Topics
- Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Protein Structure and Dynamics
- Physical Sciences › Computer Science › Computational Theory and Mathematics › Computational Drug Discovery Methods
- Physical Sciences › Materials Science › Materials Chemistry › Enzyme Structure and Function
- Keywords
- conformational sampling, protein-protein interaction, docking scoring functions, protein complex structure prediction, conformational ensembles
- MeSH terms
- Machine Learning, Protein Binding, Protein Conformation, Proteins, Molecular Docking Simulation
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
