CoChat
Sign Up
PreprintCited by 0Protein Structure and DynamicsMachine Learning in BioinformaticsComputational Drug Discovery Methods

The DOI matches its Crossref record; 1 of the 1 references it cites were found in a registry or an index.

Protein language model embedded geometric graphs power inter-protein contact prediction

Open access
Si, Yunda, Yan, Chengfei·2024·eLife
DOI 10.7554/elife.92184.2

Preprint — published as 10.7554/elife.92184.3 · Paper page

Open this paper in CoChat — free account

Abstract

Abstract Accurate prediction of contacting residue pairs between interacting proteins is very useful for structural characterization of protein-protein interactions (PPIs). Although significant improvement has been made in inter-protein contact prediction recently, there is still large room for improving the prediction accuracy. Here we present a new deep learning method referred to as PLMGraph-Inter for inter-protein contact prediction. Specifically, we employ rotationally and translationally invariant geometric graphs obtained from structures of interacting proteins to integrate multiple protein language models, which are successively transformed by graph encoders formed by geometric vector perceptrons and residual networks formed by dimensional hybrid residual blocks to predict inter-protein contacts. Extensive evaluation on multiple test sets illustrates that PLMGraph-Inter outperforms five top inter-protein contact prediction methods, including DeepHomo, GLINTER, CDPred, DeepHomo2 and DRN-1D2D_Inter by large margins. In addition, we also show that the prediction of PLMGraph-Inter can complement the result of AlphaFold-Multimer. Finally, we show leveraging the contacts predicted by PLMGraph-Inter as constraints for protein-protein docking can dramatically improve its performance for protein complex structure prediction.

Overview

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.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.7554/elife.92184.2

Cites

1 of 1 references verified0 citation errors · 0 not checkable

Statistics

Verification

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.

A few free checks a day · papers already verified open instantly · results are public

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

Flagged first, then by citations

Research works printed but not linked: 55

  1. 1998, The Cell as a Collection of Protein Machines: Preparing the Next Generation of Molecular Biologists, Cell, 92, 291, 10.1016/S0092-8674(00)80922-8 · DOI · Paper page
  2. 2016, DockQ: A Quality Measure for Protein-Protein Docking Models, PLOS ONE, 11, e0161879, 10.1371/journal.pone.0161879 · DOI · Paper page
  3. 2000, The Protein Data Bank, Nucleic Acids Research, 28, 235, 10.1093/nar/28.1.235 · DOI · Paper page
  4. 2006, Flexible protein–protein docking, Current Opinion in Structural Biology, 16, 194, 10.1016/j.sbi.2006.02.002 · DOI
  5. 2019, Protein interaction networks revealed by proteome coevolution, Science, 365, 185, 10.1126/science.aaw6718 · DOI · Paper page
  6. 2022, Robust deep learning–based protein sequence design using ProteinMPNN, Science, 0, eadd2187, 10.1126/science.add2187 · DOI · Paper page
  7. 2003, HADDOCK: A Protein−Protein Docking Approach Based on Biochemical or Biophysical Information, Journal of the American Chemical Society, 125, 1731, 10.1021/ja026939x · DOI · Paper page
  8. 2022a, Protein complex prediction with AlphaFold-Multimer, bioRxiv, 10.1101/2021.10.04.463034 · DOI · Paper page
  9. 2022b, Protein complex prediction with AlphaFold-Multimer, bioRxiv, 10.1101/2021.10.04.463034 · DOI · Paper page
  10. 2000, Structural Symmetry and Protein Function, Annual Review of Biophysics and Biomolecular Structure, 29, 105, 10.1146/annurev.biophys.29.1.105 · DOI · Paper page
  11. 2021, Large-scale discovery of protein interactions at residue resolution using co-evolution calculated from genomic sequences, Nature Communications, 12, 1, 10.1038/s41467-021-21636-z · DOI · Paper page
  12. 2022, Prediction of inter-chain distance maps of protein complexes with 2D attention-based deep neural networks, Nature Communications, 13, 1, 10.1038/s41467-022-34600-2 · DOI
  13. 2018, Accurate prediction of protein contact maps by coupling residual two-dimensional bidirectional long short-term memory with convolutional neural networks, Bioinformatics, 34, 4039, 10.1093/bioinformatics/bty481 · DOI
  14. 2021, Structural Biology in the Clouds: The WeNMR-EOSC Ecosystem, Frontiers in Molecular Biosciences, 8, 10.3389/fmolb.2021.729513 · DOI
  15. 2014, Sequence co-evolution gives 3D contacts and structures of protein complexes, eLife, 3, e03430, 10.7554/eLife.03430 · DOI · Paper page
  16. 2022a, Learning inverse folding from millions of predicted structures, bioRxiv, 10.1101/2022.04.10.487779 · DOI
  17. 2022b, Learning inverse folding from millions of predicted structures, bioRxiv, 10.1101/2022.04.10.487779 · DOI
  18. 2021. Equivariant Graph Neural Networks for 3D Macromolecular Structure. arXiv. http://arxiv.org/abs/2106.03843. · arXiv
  19. 2021, Learning from protein structure with geometric vector perceptrons, arXiv
  20. 2015, MetaPSICOV: Combining coevolution methods for accurate prediction of contacts and long range hydrogen bonding in proteins. Bioinformatics (Oxford, England, 31, 999, 10.1093/bioinformatics/btu791 · DOI
  21. 2021, CopulaNet: Learning residue co-evolution directly from multiple sequence alignment for protein structure prediction, Nature Communications, 12, 2535, 10.1038/S41467-021-22869-8 · DOI
  22. 2021, Protein–protein docking with interface residue restraints\ast, Chinese Physics B, 30, 018703, 10.1088/1674-1056/abc14e · DOI
  23. 2001, Clustering of highly homologous sequences to reduce the size of large protein databases, Bioinformatics, 17, 282, 10.1093/bioinformatics/17.3.282 · DOI
  24. 2019, ResPRE: High-accuracy protein contact prediction by coupling precision matrix with deep residual neural networks, Bioinformatics, 35, 4647, 10.1093/bioinformatics/btz291 · DOI · Paper page
  25. 2023, DeepHomo2.0: Improved protein–protein contact prediction of homodimers by transformer-enhanced deep learning, Briefings in Bioinformatics, 24, bbac499, 10.1093/bib/bbac499 · DOI
  26. 2021, Mapping, Structure and Modulation of PPI, Frontiers in Chemistry, 9, 718405, 10.3389/fchem.2021.718405 · DOI
  27. 2014, Robust and accurate prediction of residue– residue interactions across protein interfaces using evolutionary information, eLife, 3, e02030, 10.7554/eLife.02030 · DOI · Paper page
  28. 2019, Protein model quality assessment using 3D oriented convolutional neural networks, Bioinformatics, 35, 3313, 10.1093/bioinformatics/btz122 · DOI
  29. 2018, HMMER web server: 2018 update, Nucleic Acids Research, 46, W200, 10.1093/nar/gky448 · DOI · Paper page
  30. 2021, MSA Transformer, bioRxiv
  31. 2021a. MSA Transformer. Proceedings of the 38th International Conference on Machine Learning. https://proceedings.mlr.press/v139/rao21a.html.
  32. 2021b. MSA Transformer. Proceedings of the 38th International Conference on Machine Learning. https://proceedings.mlr.press/v139/rao21a.html.
  33. 2021, Biological structure and function emerge from scaling unsupervised learning to 250 million protein sequences, Proceedings of the National Academy of Sciences of the United States of America, 118, 1, 10.1073/pnas.2016239118 · DOI · Paper page
  34. 2022, A deep dilated convolutional residual network for predicting interchain contacts of protein homodimers, Bioinformatics, 38, 1904, 10.1093/bioinformatics/btac063 · DOI · Paper page
  35. 2014, CCMpred—Fast and precise prediction of protein residue-residue contacts from correlated mutations. Bioinformatics (Oxford, England, 30, 3128, 10.1093/bioinformatics/btu500 · DOI
  36. 2021, Improved protein contact prediction using dimensional hybrid residual networks and singularity enhanced loss function, Briefings in Bioinformatics, 22, bbab341, 10.1093/bib/bbab341 · DOI
  37. 2022, Protein complex structure prediction powered by multiple sequence alignments of interologs from multiple taxonomic ranks and AlphaFold2, Briefings in Bioinformatics, 23, bbac208, 10.1093/bib/bbac208 · DOI
  38. 2023, Improved inter-protein contact prediction using dimensional hybrid residual networks and protein language models. Briefings in Bioinformatics, bbad, 039, 10.1093/bib/bbad039 · DOI · Paper page
  39. 2021, D-SCRIPT translates genome to phenome with sequence-based, structure-aware, genome-scale predictions of protein-protein interactions, Cell Systems, 12, 969, 10.1016/j.cels.2021.08.010 · DOI
  40. 2003, Protein complexes and functional modules in molecular networks, Proceedings of the National Academy of Sciences, 100, 12123, 10.1073/pnas.2032324100 · DOI
  41. 2019, HH-suite3 for fast remote homology detection and deep protein annotation, BMC Bioinformatics, 20, 1, 10.1186/s12859-019-3019-7 · DOI · Paper page
  42. 2018, Clustering huge protein sequence sets in linear time, Nature Communications, 9, 1, 10.1038/s41467-018-04964-5 · DOI · Paper page
  43. 2020, Review of multimer protein–protein interaction complex topology and structure prediction\ast, Chinese Physics B, 29, 108707, 10.1088/1674-1056/abb659 · DOI
  44. 2017, Large-scale identification of coevolution signals across homo-oligomeric protein interfaces by direct coupling analysis, Proceedings of the National Academy of Sciences, 114, E2662, 10.1073/pnas.1615068114 · DOI
  45. 2016, The HADDOCK2.2 Web Server: User-Friendly Integrative Modeling of Biomolecular Complexes, Journal of Molecular Biology, 428, 720, 10.1016/j.jmb.2015.09.014 · DOI · Paper page
  46. 2017, Attention is All you Need, Advances in Neural Information Processing Systems, 30
  47. 2017, Accurate De Novo Prediction of Protein Contact Map by Ultra-Deep Learning Model, PLOS Computational Biology, 13, 1005324, 10.1371/journal.pcbi.1005324 · DOI · Paper page
  48. 2009, Identification of direct residue contacts in protein–protein interaction by message passing, Proceedings of the National Academy of Sciences, 106, 67, 10.1073/pnas.0805923106 · DOI · Paper page
  49. 2023, Integration of pre-trained protein language models into geometric deep learning networks, Communications Biology, 6, 1, 10.1038/s42003-023-05133-1 · DOI
  50. 2022, Inter-chain contact map prediction for protein complex based on graph attention network and triangular multiplication update, 2022 IEEE International Conference on Bioinformatics and Biomedicine (BIBM, 2143, 10.1109/BIBM55620.2022.9995360 · DOI
  51. 2022, Deep graph learning of inter-protein contacts, Bioinformatics, 38, 947, 10.1093/bioinformatics/btab761 · DOI · Paper page
  52. 2021, Accurate prediction of inter-protein residue-residue contacts for homo-oligomeric protein complexes, Briefings in Bioinformatics, 22, 1, 10.1093/bib/bbab038 · DOI
  53. 2018, ComplexContact: A web server for inter-protein contact prediction using deep learning, Nucleic Acids Research, 46, W432, 10.1093/nar/gky420 · DOI
  54. 2004, Scoring function for automated assessment of protein structure template quality. Proteins: Structure, Function, and Bioinformatics, 57, 702, 10.1002/prot.20264 · DOI · Paper page
  55. 2005, TM-align: A protein structure alignment algorithm based on the TM-score, Nucleic Acids Research, 33, 2302, 10.1093/nar/gki524 · DOI · Paper page

Metadata

Authors
Si, Y., & Yan, C.
Affiliations
Huazhong University of Science and Technology (CN)
Corresponding author
Yan, Chengfei
ORCID
Venue
eLife
Publisher
eLife Sciences Publications, Ltd
Published
2024-03-14
ISSN
2050-084X (ISSN-L)
DOI
10.7554/elife.92184.2
OpenAlex
W4392810182
Cited by
0
References
1
Type
posted content
Language
en
Open access
Licence
Locations
eLife (acceptedVersion, cc-by)
Funders
  • National Natural Science Foundation of China (32101001)
  • Huazhong University of Science and Technology (3004012167)
Topics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Protein Structure and Dynamics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Machine Learning in Bioinformatics
  • Life Sciences › Pharmacology, Toxicology and Pharmaceutics › Drug Discovery › Computational Drug Discovery Methods
Keywords
protein-protein interaction, protein language models, geometric graphs, graph neural networks, protein complex structure prediction, protein-protein docking, AlphaFold-Multimer

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