Machine learning solutions for predicting protein–protein interactions
Open accessThe DOI matches its Crossref record; 21 of the 21 references it cites were found in a registry or an index.
Abstract
Proteins are “social molecules.” Recent experimental evidence supports the notion that large protein aggregates, known as biomolecular condensates, affect structurally and functionally many biological processes. Condensate formation may be permanent and/or time dependent, suggesting that biological processes can occur locally, depending on the cell needs. The question then arises as to which extent we can monitor protein‐aggregate formation, both experimentally and theoretically and then predict/simulate functional aggregate formation. Available data are relative to mesoscopic interacting networks at a proteome level, to protein‐binding affinity data, and to interacting protein complexes, solved with atomic resolution. Powerful algorithms based on machine learning (ML) can extract information from data sets and infer properties of never‐seen‐before examples. ML tools address the problem of protein–protein interactions (PPIs) adopting different data sets, input features, and architectures. According to recent publications, deep learning is the most successful method. However, in ML‐computational biology, convincing evidence of a success story comes out by performing general benchmarks on blind data sets. Results indicate that the state‐of‐the‐art ML approaches, based on traditional and/or deep learning, can still be ameliorated, irrespectively of the power of the method and richness in input features. This being the case, it is quite evident that powerful methods still are not trained on the whole possible spectrum of PPIs and that more investigations are necessary to complete our knowledge of PPI‐functional interactions. This article is categorized under: Software > Molecular Modeling Structure and Mechanism > Computational Biochemistry and Biophysics Data Science > Artificial Intelligence/Machine Learning Molecular and Statistical Mechanics > Molecular 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
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
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 1×
“115 Furthermore, other papers still in bioRxiv, support the notion that the problem of predicting the interfaces of protein complexes, can take advantage of the deep learning-based methods included in AlphaFold2.”
BackgroundGeometric Deep Learning: Going beyond Euclidean data
Bronstein, M. M., Bruna, J., LeCun, Y., Szlam, A., & Vandergheynst, P. (2017) · IEEE Signal Processing Magazine · doi:10.1109/msp.2017.2693418 · Cited by 3,661 · cited 1×
“99 These approaches are useful for modeling data that cannot be easily represented into a standard Euclidean space, that is, data having an underlying non-Euclidean structure such as graphs or networks.”
BackgroundHighly accurate protein structure prediction for the human proteome
Tunyasuvunakool, K., Adler, J., Wu, Z., Green, T., Zielinski, M., Žídek, A., Bridgland, A., Cowie, A., Meyer, C., Laydon, A., et al. (2021) · Nature · doi:10.1038/s41586-021-03828-1 · Cited by 3,025
A guide to machine learning for biologists
Greener, J. G., Kandathil, S. M., Moffat, L., & Jones, D. T. (2022) · Nature Reviews Molecular Cell Biology · doi:10.1038/s41580-021-00407-0 · Cited by 2,127 · cited 1×
“Methods referred to as traditional in the recent literature, 15 include shallow neural networks (NN), support vector machines (SVMs), random forests (RFs), and probabilistic methods such as Hidden Markov Models (HMMs) and conditional random fields (CRFs).”
MethodImproved prediction of protein-protein interactions using AlphaFold2
Bryant, P., Pozzati, G., & Elofsson, A. (2021) · doi:10.1101/2021.09.15.460468 · Cited by 1,022 · cited 3×
“Docking of protein models can be improved by adopting AlphaFold2 and a docking method (ClusPro)(117) and heterodimeric protein complex interactions can be better predicted by including AlphaFold2.(118) All this work suggests that indeed the computational power of deep learning developments can efficiently extract info…”
MethodA framework for understanding the functions of biomolecular condensates across scales
Lyon, A. S., Peeples, W. B., & Rosen, M. K. (2021) · Nature Reviews Molecular Cell Biology · doi:10.1038/s41580-020-00303-z · Cited by 895 · cited 6×
“The second definition is based on the computation of inter-residue distances: interface residues are those having at least one residue of another subunit at a distance below a defined threshold (routinely between 5 and 8 Å).”
BackgroundInfluentialLu, 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 · cited 1×
“9 In this dynamic scenario, the problem of protein – protein molecular interactions (PPI) is evidently an issue.”
BackgroundSAbDab: 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
Ovchinnikov, S., Kamisetty, H., & Baker, D. (2014) · eLife · doi:10.7554/elife.02030 · Cited by 629 · cited 1×
“71 A subset of sequence-based methods, 72,73 routinely partner-specific approaches that identify pairs of interacting residues between two input partners, are based on the analysis of protein coevolution.”
BackgroundModeling the language of life – Deep Learning Protein Sequences
Heinzinger, M., Elnaggar, A., Wang, Y., Dallago, C., Nechaev, D., Matthes, F., & Rost, B. (2019) · doi:10.1101/614313 · Cited by 606
Modeling aspects of the language of life through transfer-learning protein sequences
Heinzinger, M., Elnaggar, A., Wang, Y., Dallago, C., Nechaev, D., Matthes, F., & Rost, B. (2019) · BMC Bioinformatics · doi:10.1186/s12859-019-3220-8 · Cited by 606 · cited 1×
“65,66 Some of these approaches have been recently adopted also in the field of PPI site prediction.”
BackgroundSingle-cell multiomics: technologies and data analysis methods
Lee, J., Hyeon, D. Y., & Hwang, D. (2020) · Experimental & Molecular Medicine · doi:10.1038/s12276-020-0420-2 · Cited by 524
Sequence co-evolution gives 3D contacts and structures of protein complexes
Hopf, T. A., Schärfe, C. P. I., Rodrigues, J. P. G. L. M., Green, A. G., Kohlbacher, O., Sander, C., Bonvin, A. M. J. J., & Marks, D. S. (2014) · eLife · doi:10.7554/elife.03430 · Cited by 515 · cited 1×
“71 A subset of sequence-based methods, 72,73 routinely partner-specific approaches that identify pairs of interacting residues between two input partners, are based on the analysis of protein coevolution.”
BackgroundVreven, T., Moal, I. H., Vangone, A., Pierce, B. G., Kastritis, P. L., Torchala, M., Chaleil, R., Jiménez-García, B., Bates, P. A., Fernandez-Recio, J., et al. (2015) · Journal of Molecular Biology · doi:10.1016/j.jmb.2015.07.016 · Cited by 424
Jankauskaitė, J., Jiménez-García, B., Dapkūnas, J., Fernández-Recio, J., & Moal, I. H. (2019) · Bioinformatics · doi:10.1093/bioinformatics/bty635 · Cited by 389 · cited 1×
“Specific and derived databases organize structures according to given properties, 39 like ProtCID, a data resource for structural information on protein interactions.”
BackgroundAffinity-purification coupled to mass spectrometry: basic principles and strategies
Dunham, W. H., Mullin, M., & Gingras, A. (2012) · PROTEOMICS · doi:10.1002/pmic.201100523 · Cited by 368 · cited 1×
“31 Affinity purification is a chromatographic assay for isolating all the interactors of a bait protein from a mixture 32 while in co-fractionation experiments native complexes are separated with physicochemical techniques.”
MethodDOME: recommendations for supervised machine learning validation in biology
Walsh, I., Fishman, D., Garcia-Gasulla, D., Titma, T., Pollastri, G., ELIXIR Machine Learning Focus Group, Capriotti, E., Casadio, R., Capella-Gutierrez, S., Cirillo, D., et al. (2021) · Nature Methods · doi:10.1038/s41592-021-01205-4 · Cited by 282
Self-citationThe binary protein-protein interaction landscape of Escherichia coli
Rajagopala, S. V., Sikorski, P., Kumar, A., Mosca, R., Vlasblom, J., Arnold, R., Franca-Koh, J., Pakala, S. B., Phanse, S., Ceol, A., et al. (2014) · Nature Biotechnology · doi:10.1038/nbt.2831 · Cited by 233
Computational prediction of protein–protein binding affinities
Siebenmorgen, T., & Zacharias, M. (2020) · WIREs Computational Molecular Science · doi:10.1002/wcms.1448 · Cited by 184 · cited 2×
“Supervised machine-learning is applied for developing protein-targeted scoring functions for the prediction of binding affinity 41,42 (for an extensive description of recent docking methods, see Reference 40 and references therein).”
MethodThe structural coverage of the human proteome before and after AlphaFold
Porta-Pardo, E., Ruiz-Serra, V., Valentini, S., & Valencia, A. (2022) · PLOS Computational Biology · doi:10.1371/journal.pcbi.1009818 · Cited by 131 · cited 1×
“After AlphaFold2 models, the fraction of the dark structural proteome decreased from 26% to 10%, allowing a coverage increase of the critically important sets of disease-associated genes and mutations.(114) One method(115) integrates information from three different levels, including protein sequence, distance map, an…”
Background
Research works printed but not linked: 100
- Kessel A, 2018, Introduction to proteins: structure, function, and motion. Mathematical and computational biology series, 932, 10.1201/9781315113876 · DOI
- 10.1038/nrm.2017.93 · doi:10.1038/nrm.2017.93
- 10.1016/j.tcb.2018.02.004 · DOI · Paper page
- 10.1016/j.jmb.2018.08.003 · DOI · Paper page
- 10.1038/s41580-020-00326-6 · DOI · Paper page
- 10.1038/s41580-020-0272-6 · DOI · Paper page
- 10.1038/s41580-020-0264-6 · DOI · Paper page
- 10.1021/acs.biochem.1c00376 · DOI
- 10.1016/j.cbpa.2018.11.003 · DOI
- 10.1039/D1CS00548K · DOI · Paper page
- Baldi P, 2021, Deep learning in science, 371, 10.1017/9781108955652 · DOI
- Bishop CM, 2006, Pattern recognition and machine learning. Information science and statistics, 738
- Goodfellow I, 2016, Deep learning. Adaptive computation and machine learning, 775
- 10.1007/978-1-0716-1641-3_16 · DOI
- 10.1038/s41580-019-0176-5 · DOI
- 10.1002/prot.26222 · DOI · Paper page
- 10.1146/annurev-biodatasci-011720-104428 · DOI
- 10.1186/s13578-021-00691-5 · DOI
- 10.2174/0929867325666180214113704 · DOI
- 10.1038/s41598-019-46369-4 · DOI · Paper page
- 10.1186/s12859-020-03646-8 · DOI · Paper page
- 10.1038/s41467-019-09177-y · DOI · Paper page
- 10.3389/fgene.2020.00291 · DOI
- 10.1002/cpps.70 · DOI
- 10.1016/j.xpro.2021.100370 · DOI
- 10.1007/s00018-021-03856-0 · DOI
- 10.1016/j.sbi.2013.08.002 · DOI
- 10.1016/j.tibs.2014.10.005 · DOI · Paper page
- 10.1039/9781788016544-00001 · DOI
- 10.1021/acs.jcim.8b00545 · DOI · Paper page
- 10.1007/978-1-4939-9752-7_16 · DOI
- 10.2174/0929867327666200515101820 · doi:10.2174/0929867327666200515101820
- 10.1038/s41467-020-14301-4 · DOI · Paper page
- 10.1017/S0033583508004708 · DOI · Paper page
- 10.1038/s41592-019-0666-6 · DOI · Paper page
- 10.1021/acs.jcim.7b00028 · DOI · Paper page
- 10.1093/bioinformatics/bty573 · DOI · Paper page
- 10.1093/bioinformatics/btv077 · DOI · Paper page
- 10.1093/bioinformatics/17.5.455 · DOI · Paper page
- 10.1371/journal.pone.0125811 · DOI · Paper page
- 10.1186/s12859-017-1700-2 · DOI · Paper page
- 10.1016/j.neucom.2018.02.097 · DOI
- 10.3389/fbioe.2020.00390 · DOI
- 10.1021/pr100618t · DOI · Paper page
- 10.1093/nar/gkn159 · DOI · Paper page
- 10.1109/JBHI.2018.2845866 · DOI
- 10.1142/S021972000500120X · DOI · Paper page
- 10.1038/nmeth.2259 · DOI · Paper page
- 10.1371/journal.pone.0031826 · DOI · Paper page
- 10.1093/bioinformatics/btaa750 · DOI · Paper page
- 10.1093/bioinformatics/btz428 · DOI
- 10.1093/bioinformatics/btz324 · DOI · Paper page
- 10.1093/bioinformatics/bty647 · DOI · Paper page
- 10.1093/bioinformatics/btq302 · DOI · Paper page
- 10.1093/bioinformatics/bty178 · DOI · Paper page
- 10.1016/j.jmb.2020.02.026 · DOI · Paper page
- 10.1109/TCBB.2019.2953908 · DOI
- 10.1093/nar/28.1.374 · DOI · Paper page
- 10.1007/BF01025492 · DOI · Paper page
- 10.1016/j.neucom.2016.02.022 · DOI
- Zeng M, 2020, Protein–protein interaction site prediction through combining local and global features with deep neural networks, Bioinformatics, 36, 1114, 10.1093/bioinformatics/btz699 · DOI · Paper page
- 10.1016/j.jtbi.2014.01.028 · DOI
- 10.1002/prot.20593 · DOI
- Hou Q, 2017, Seeing the trees through the forest: sequence‐based homo‐ and heteromeric protein‐protein interaction sites prediction using random forest, Bioinformatics, 33, 1479, 10.1093/bioinformatics/btx005 · DOI
- 10.1186/s12859-015-0758-y · DOI
- 10.1002/prot.22830 · DOI · Paper page
- 10.1002/prot.22106 · DOI · Paper page
- Savojardo C, 2017, ISPRED4: interaction sites PREDiction in protein structures with a refining grammar model, Bioinformatics, 33, 1656, 10.1093/bioinformatics/btx044 · DOI
- 10.1186/1471-2105-15-82 · DOI
- 10.1016/S0006-3495(03)75060-7 · DOI
- 10.1093/bioinformatics/18.7.980 · DOI · Paper page
- Koenderink JJ, 1990, Solid shape. Artificial intelligence, 699
- 10.1002/bip.360330205 · DOI
- 10.1186/s12859-018-2043-3 · DOI
- 10.1016/j.csbj.2020.11.051 · DOI
- 10.1073/pnas.0906146106 · DOI · Paper page
- 10.1002/prot.23172 · DOI
- 10.1002/prot.21248 · DOI · Paper page
- 10.1093/bioinformatics/btx585 · DOI
- 10.1186/1471-2105-10-381 · DOI
- 10.1186/1471-2105-15-277 · DOI
- 10.1093/bioinformatics/btl660 · DOI
- 10.1093/bioinformatics/btab643 · DOI · Paper page
- 10.1093/nar/gkaa1058 · DOI · Paper page
- 10.1093/nar/gku397 · DOI · Paper page
- 10.1093/bioinformatics/btu626 · DOI · Paper page
- 10.1016/j.jmb.2007.05.022 · DOI · Paper page
- 10.1186/1471-2105-12-352 · DOI
- 10.1093/bioinformatics/btr638 · DOI · Paper page
- 10.1186/s12859-017-1921-4 · doi:10.1186/s12859-017-1921-4
- 10.1186/1471-2105-13-41 · DOI · Paper page
- 10.1021/ci8000315 · DOI
- 10.1073/pnas.1005894107 · DOI · Paper page
- 10.1016/j.sbi.2021.01.008 · DOI · Paper page
- 10.1039/C9CP06554G · DOI · Paper page
- 10.1016/j.cels.2021.08.010 · DOI · Paper page
- 10.1126/science.abj8754 · doi:10.1126/science.abj8754 · Paper page
- 10.1101/2021.10.04.463034v1 · doi:10.1101/2021.10.04.463034v1
- 10.1101/2021.09.07.459290v1 · doi:10.1101/2021.09.07.459290v1
- 10.1038/s41467-022-28865-w · doi:10.1038/s41467-022-28865-w · Paper page
Cited by
(2023) · doi:10.1093/bioadv/vbad191 · Cited by 37
Zhou, Z., Yin, Y., Han, H., Jia, Y., Koh, J. H., Kong, A. W.-K., & Mu, Y. (2024) · Journal of Chemical Information and Modeling · doi:10.1021/acs.jcim.4c01850 · Cited by 30
Seq-InSite: sequence supersedes structure for protein interaction site prediction
Hosseini, S., Golding, G. B., & Ilie, L. (2024) · Bioinformatics · doi:10.1093/bioinformatics/btad738 · Cited by 26
(2024) · doi:10.3390/biomimetics9030162 · Cited by 7
Metadata
- Authors
- Casadio, R., Martelli, P. L., & Savojardo, C.
- Venue
- WIREs Computational Molecular Science
- Publisher
- Wiley
- Published
- 2022-03-29
- Volume
- 12
- Issue
- 6
- ISSN
- 1759-0876 (ISSN-L)
- 1759-0884
- OpenAlex
- W4220933909
- Cited by
- 80
- References
- 21
- FWCI
- 4.83
- Type
- journal article
- Language
- en
- Open access
- hybrid
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.