CoChat
Sign Up
PreprintCited by 84Machine Learning in BioinformaticsProtein Structure and DynamicsComputational Drug Discovery Methods

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

EquiPNAS: improved protein-nucleic acid binding site prediction using protein-language-model-informed equivariant deep graph neural networks

Open access
Roche, Rahmatullah, Moussad, Bernard, Shuvo, Md Hossain, Tarafder, Sumit, Bhattacharya, Debswapna·2023·Nucleic Acids Research
DOI 10.1101/2023.09.14.557719PMC PMC10515942

Preprint — published as 10.1093/nar/gkae039 · Paper page

Open this paper in CoChat — free accountOpen PDF

Abstract

Protein language models (pLMs) trained on a large corpus of protein sequences have shown unprecedented scalability and broad generalizability in a wide range of predictive modeling tasks, but their power has not yet been harnessed for predicting protein-nucleic acid binding sites, critical for characterizing the interactions between proteins and nucleic acids. Here we present EquiPNAS, a new pLM-informed E(3) equivariant deep graph neural network framework for improved protein-nucleic acid binding site prediction. By combining the strengths of pLM and symmetry-aware deep graph learning, EquiPNAS consistently outperforms the state-of-the-art methods for both protein-DNA and protein-RNA binding site prediction on multiple datasets across a diverse set of predictive modeling scenarios ranging from using experimental input to AlphaFold2 predictions. Our ablation study reveals that the pLM embeddings used in EquiPNAS are sufficiently powerful to dramatically reduce the dependence on the availability of evolutionary information without compromising on accuracy, and that the symmetry-aware nature of the E(3) equivariant graph-based neural architecture offers remarkable robustness and performance resilience. EquiPNAS is freely available at https://github.com/Bhattacharya-Lab/EquiPNAS .

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 · 10.1101/2023.09.14.557719

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: 50

  1. 10.1038/nature07348 · DOI
  2. 10.1093/nar/gkq617 · DOI
  3. 10.1093/bioinformatics/btz324 · DOI
  4. 10.1093/bioinformatics/btq295 · DOI
  5. 2013, Structure-based function prediction of uncharacterized protein using binding sites comparison, PLoS computational biology, 9, e1003341, 10.1371/journal.pcbi.1003341 · DOI
  6. 2010, Understanding and predicting druggability, A high-throughput method for detection of drug binding sites. Journal of medicinal chemistry, 53, 5858
  7. 2019, Improving the prediction of protein–nucleic acids binding residues via multiple sequence profiles and the consensus of complementary methods, Bioinformatics, 35, 930, 10.1093/bioinformatics/bty756 · DOI
  8. 2021, NCBRPred: predicting nucleic acid binding residues in proteins based on multilabel learning, Briefings in bioinformatics, 22, bbaa397, 10.1093/bib/bbaa397 · DOI
  9. 10.1021/acs.jcim.8b00749 · DOI
  10. 10.1093/bib/bbab336 · DOI
  11. 10.1371/journal.pone.0097725 · DOI · Paper page
  12. 10.1006/jmbi.2000.4474 · DOI
  13. 2016, Predicting protein-DNA binding residues by weightedly combining sequence-based features and boosting multiple SVMs, IEEE/ACM transactions on computational biology and bioinformatics, 14, 1389
  14. 2013, Designing template-free predictor for targeting protein-ligand binding sites with classifier ensemble and spatial clustering, IEEE/ACM transactions on computational biology and bioinformatics, 10, 994, 10.1109/TCBB.2013.104 · DOI
  15. 10.1093/nar/gky439 · DOI
  16. 2013, DNABind: A hybrid algorithm for structure-based prediction of DNA-binding residues by combining machine learning-and template-based approaches, PROTEINS: structure, Function, and Bioinformatics, 81, 1885, 10.1002/prot.24330 · DOI
  17. 10.1093/bioinformatics/btx350 · DOI
  18. 10.1093/nar/gku681 · DOI
  19. 2019, A deep learning framework to predict binding preference of RNA constituents on protein surface, Nature communications, 10, 4941, 10.1038/s41467-019-12920-0 · DOI
  20. 2021, GraphBind: protein structural context embedded rules learned by hierarchical graph neural networks for recognizing nucleicacid-binding residues, Nucleic acids research, 49, e51, 10.1093/nar/gkab044 · DOI
  21. 2022, AlphaFold2-aware protein–DNA binding site prediction using graph transformer, Briefings in Bioinformatics, 23, bbab564, 10.1093/bib/bbab564 · DOI · Paper page
  22. 1971, Protein data bank, Nature New Biol, 233, 223, 10.1038/newbio233223b0 · DOI
  23. 10.1093/NAR/GKAB1061 · DOI · Paper page
  24. 2007, ProtTrans: Towards cracking the language of Life’s code through selfsupervised deep learning and high performance computing, arXiv 2020. arXiv preprint
  25. 2022, ProteinBERT: a universal deep-learning model of protein sequence and function, Bioinformatics, 38, 2102, 10.1093/bioinformatics/btac020 · DOI · Paper page
  26. 2022, ProtGPT2 is a deep unsupervised language model for protein design, Nature communications, 13, 4348, 10.1038/s41467-022-32007-7 · DOI · Paper page
  27. 2022, Single-sequence protein structure prediction using a language model and deep learning, Nature Biotechnology, 40, 1617, 10.1038/s41587-022-01432-w · DOI · Paper page
  28. 10.1073/pnas.2016239118 · DOI · Paper page
  29. 10.1126/science.ade2574 · DOI · Paper page
  30. Vaswani, A. et al. Attention is all you need. Advances in neural information processing systems 30 (2017).
  31. Madani, A. et al. Large language models generate functional protein sequences across diverse families. Nature Biotechnology, 1–8 (2023). · DOI · Paper page
  32. 10.1021/acs.iecr.1c04943 · DOI
  33. 2021, in Proceedings of the 38th International Conference on Machine Learning, 139, 9323
  34. 2023, E (3) equivariant graph neural networks for robust and accurate protein-protein interaction site prediction, PLOS Computational Biology, 19, e1011435, 10.1371/journal.pcbi.1011435 · DOI
  35. 10.1073/pnas.2303499120 · DOI
  36. Hie, B. L. et al. Efficient evolution of human antibodies from general protein language models. Nature Biotechnology (2023). · DOI
  37. 2023, Integration of pre-trained protein language models into geometric deep learning networks, Communications Biology, 6, 876, 10.1038/s42003-023-05133-1 · DOI
  38. 10.1093/nar/25.17.3389 · DOI · Paper page
  39. 10.1038/s41592-022-01488-1 · DOI · Paper page
  40. 10.1038/nbt.3988 · DOI · Paper page
  41. 2020, Learning from Protein Structure with Geometric Vector Perceptrons
  42. 2019, PyTorch: An Imperative Style, High-Performance Deep Learning Library
  43. 2019, Deep Graph Library: A Graph-Centric, Highly-Performant Package for Graph Neural Networks
  44. 2016, SGDR: Stochastic Gradient Descent with Warm Restarts
  45. 2014, Adam: A Method for Stochastic Optimization
  46. 10.1093/nar/gks966 · DOI
  47. 10.1093/bioinformatics/btq003 · DOI
  48. 10.1093/bioinformatics/16.4.404 · DOI
  49. 10.1038/nmeth.1818 · DOI · Paper page
  50. 2014, Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translation

Metadata

Authors
Roche, R., Moussad, B., Shuvo, M. H., Tarafder, S., & Bhattacharya, D.
Affiliations
Virginia Tech (US)
Corresponding author
Bhattacharya, Debswapna
ORCID
Venue
Nucleic Acids Research
Publisher
openRxiv
Published
2023-09-16
Volume
52
Issue
5
Pages
e27-e27
ISSN
0305-1048 (ISSN-L)
DOI
10.1101/2023.09.14.557719
PMC
PMC10515942
OpenAlex
W4391291188
Cited by
84
References
1
Citations by year
2026: 25 · 2025: 45 · 2024: 14
FWCI
11.44
Type
journal article
Language
en
Open access
Licence
cc-by-nc
Locations
  • Nucleic Acids Research (publishedVersion, cc-by-nc)
  • bioRxiv (Cold Spring Harbor Laboratory) (acceptedVersion, cc-by-nd)
  • PubMed (publishedVersion)
  • PubMed Central (submittedVersion, cc-by-nd)
Funders
  • National Science Foundation (2208679)
  • National Institute of General Medical Sciences (R35GM138146)
Topics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Machine Learning in Bioinformatics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Protein Structure and Dynamics
  • Life Sciences › Pharmacology, Toxicology and Pharmaceutics › Drug Discovery › Computational Drug Discovery Methods
Keywords
protein language models, protein-DNA interactions, RNA-protein interactions, protein structure prediction, AlphaFold2, evolutionary information

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