A comparison of antibody–antigen complex sequence‐to‐structure prediction methods and their systematic biases
Open accessThe DOI matches its Crossref record; 13 of the 13 references it cites were found in a registry or an index.
Abstract
The ability to accurately predict antibody–antigen complex structures from their sequences could greatly advance our understanding of the immune system and would aid in the development of novel antibody therapeutics. There have been considerable recent advancements in predicting protein–protein interactions (PPIs) fueled by progress in machine learning (ML). To understand the current state of the field, we compare six representative methods for predicting antibody–antigen complexes from sequence, including two deep learning approaches trained to predict PPIs in general (AlphaFold‐Multimer and RoseTTAFold), two composite methods that initially predict antibody and antigen structures separately and dock them (using antibody‐mode ClusPro), local refinement in Rosetta (SnugDock) of globally docked poses from ClusPro, and a pipeline combining homology modeling with rigid‐body docking informed by ML‐based epitope and paratope prediction (AbAdapt). We find that AlphaFold‐Multimer outperformed other methods, although the absolute performance leaves considerable room for improvement. AlphaFold‐Multimer models of lower quality display significant structural biases at the level of tertiary motifs (TERMs) toward having fewer structural matches in non‐antibody‐containing structures from the Protein Data Bank (PDB). Specifically, better models exhibit more common PDB‐like TERMs at the antibody–antigen interface than worse ones. Importantly, the clear relationship between performance and the commonness of interfacial TERMs suggests that the scarcity of interfacial geometry data in the structural database may currently limit the application of ML to the prediction of antibody–antigen interactions.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
ColabFold - Making protein folding accessible to all
Mirdita, M., Schütze, K., Moriwaki, Y., Heo, L., Ovchinnikov, S., & Steinegger, M. (2021) · doi:10.1101/2021.08.15.456425 · Cited by 9,391
Accurate prediction of protein structures and interactions using a 3-track network
Baek, M., DiMaio, F., Anishchenko, I., Dauparas, J., Ovchinnikov, S., Lee, G. R., Wang, J., Cong, Q., Kinch, L. N., Schaeffer, R. D., et al. (2021) · doi:10.1101/2021.06.14.448402 · Cited by 5,549 · cited 2×
“RoseTTAFold models were generated using the Robetta webserver (Baek et al., 2021; Hiranuma et al., 2020) in February 2023.”
Protein complex prediction with AlphaFold-Multimer
Evans, R., O’Neill, M., Pritzel, A., Antropova, N., Senior, A., Green, T., Žídek, A., Bates, R., Blackwell, S., Yim, J., et al. (2021) · doi:10.1101/2021.10.04.463034 · Cited by 3,808 · cited 3×
“The number of effective sequences for MSAs (N eff ) was calculated using a sequence cutoff of 80%, as previously performed in evaluating AlphaFold (Evans et al., 2022).”
InfluentialHarnessing protein folding neural networks for peptide–protein docking
Tsaban, T., Varga, J. K., Avraham, O., Ben-Aharon, Z., Khramushin, A., & Schueler-Furman, O. (2022) · Nature Communications · doi:10.1038/s41467-021-27838-9 · Cited by 1,205
SAbDab: the structural antibody database
Dunbar, J., Krawczyk, K., Leem, J., Baker, T., Fuchs, A., Georges, G., Shi, J., & Deane, C. M. (2014) · Nucleic Acids Research · doi:10.1093/nar/gkt1043 · Cited by 674
DockQ: A Quality Measure for Protein-Protein Docking Models
Basu, S., & Wallner, B. (2016) · PLOS ONE · doi:10.1371/journal.pone.0161879 · Cited by 460 · cited 2×
“A DOCKQ score ≥ 0.23 is commonly used as a cutoff to differentiate correct and incorrect structures (Basu & Wallner, 2016), and that cutoff is used here.”
InfluentialIlluminating protein space with a programmable generative model
Ingraham, J. B., Baranov, M., Costello, Z., Barber, K. W., Wang, W., Ismail, A., Frappier, V., Lord, D. M., Ng-Thow-Hing, C., Van Vlack, E. R., et al. (2023) · Nature · doi:10.1038/s41586-023-06728-8 · Cited by 444 · cited 1×
“Diffusion-based methods such as Chroma (Ingraham et al., 2023) and DiffDock-PP (Ketata et al., 2023) offer a novel architecture for predicting protein complexes.”
Self-citationImproved protein structure refinement guided by deep learning based accuracy estimation
Hiranuma, N., Park, H., Baek, M., Anishchenko, I., Dauparas, J., & Baker, D. (2021) · Nature Communications · doi:10.1038/s41467-021-21511-x · Cited by 243 · cited 1×
“RoseTTAFold models were generated using the Robetta webserver (Baek et al., 2021; Hiranuma et al., 2020) in February 2023.”
Yin, R., & Pierce, B. G. (2024) · Protein Science · doi:10.1002/pro.4865 · Cited by 145
The H3 loop of antibodies shows unique structural characteristics
Regep, C., Georges, G., Shi, J., Popovic, B., & Deane, C. M. (2017) · Proteins Structure Function and Bioinformatics · doi:10.1002/prot.25291 · Cited by 99 · cited 1×
“…– antigen complexes, but has been evaluated in its ability to predict antibodies on their own — in which case it predicts the H3 loop crucial for most binding motifs (Regep et al., 2017) better than ABodyBuilder and comparably to SWISS-MODEL, but the overall 3D structure worse (Liang et al., 2022).”
Cited by
Wu, D., Yin, R., Chen, G., Ribeiro-Filho, H. V., Cheung, M., Robbins, P. F., Mariuzza, R. A., & Pierce, B. G. (2024) · Science Advances · doi:10.1126/sciadv.adq6150 · Cited by 27
Evaluating deep learning based structure prediction methods on antibody–antigen complexes
Fromm, S., Ludaic, M., & Elofsson, A. (2026) · Bioinformatics · doi:10.1093/bioinformatics/btag136 · Cited by 19
Evaluation of AlphaFold modeling for elucidation of nanobody–peptide epitope interactions
Sachdev, S., Roy, S., Saha, S. J., Zhao, G., Kumariya, R., Creemer, B. A., Yin, R., Pierce, B. G., Bewley, C. A., & Cheloha, R. W. (2025) · Journal of Biological Chemistry · doi:10.1016/j.jbc.2025.110268 · Cited by 8
Deep geometric framework to predict antibody–antigen binding affinity
(2025) · doi:10.1016/j.jsb.2025.108257 · Cited by 3
Metadata
- Authors
- McCoy, K. M., Ackerman, M. E., & Grigoryan, G.
- Venue
- Protein Science
- Publisher
- Wiley
- Published
- 2024-08-21
- Volume
- 33
- Issue
- 9
- Pages
- e5127-e5127
- ISSN
- 0961-8368 (ISSN-L)
- 1469-896X
- OpenAlex
- W4401725571
- Cited by
- 25
- References
- 13
- FWCI
- 2.90
- Type
- journal article
- Language
- en
- Open access
- green
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
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.