Abstract
Genotype-phenotype relationships are mediated through intricate networks of physical and functional interactions among macromolecules. Knowledge of the interactome is vital to understand and model genetics and cellular biology. Recent advances in accurately predicting tertiary protein structures using artificial intelligence (AI) approaches such as AlphaFold1 have revived the vision that the protein-protein interactome might be fully predictable through computational modeling of quaternary structures. Here we present a comprehensive experimental framework to systematically assess the impact of AI-driven interactome predictions for yeast2 and human3. We find that the quality of high-confidence predictions is on par with established experimental approaches. However, in proteome-wide screening, the tested AI approaches underperform in the discovery of strictly novel protein-protein interactions (PPIs) compared to experimental reference interactome maps. In particular, the yeast interactome map described here identifies >40-fold more novel PPIs than its AI counterpart. Strikingly, AlphaFold provides structural models for a substantial number of experimentally identified PPIs missed by the virtual screens. Our results suggest that, at this stage, the main contribution of AI predictions is to provide quaternary structure models for experimentally identified PPIs.
Data availability
Protein interaction data have been submitted to the IMEx consortium (http://www.imexconsortium.org) through IntAct57 and assigned the identifier IM-30553. Predicted structures of YeRI PPIs are deposited at https://doi.org/10.5281/zenodo.18601049. YeRI, Y2H-union-25, and ValBin-25 maps are available at our OpenPIP84 website: https://yeast.interactome-atlas.org/. Source data are provided with this paper.
Code availability
Analysis code is available at https://github.com/ccsb-dfci/ai-interactome-experimental-assessment, archived together with the input data at https://doi.org/10.5281/zenodo.18499797.
References
Jumper, J. et al. Highly accurate protein structure prediction with AlphaFold. Nature 596, 583–589 (2021).
Humphreys, I. R. et al. Computed structures of core eukaryotic protein complexes. Science 374, eabm4805 (2021).
Zhang, J. et al. Predicting protein-protein interactions in the human proteome. Science 390, eadt1630 (2025).
Luck, K., Sheynkman, G. M., Zhang, I. & Vidal, M. Proteome-scale human interactomics. Trends Biochem. Sci. 42, 342–354 (2017).
Evans, R., O’Neill, M., Pritzel, A., Antropova, N. & Senior, A. W. Protein complex prediction with AlphaFold-multimer. Preprint at bioRxiv https://doi.org/10.1101/2021.10.04.463034 (2021).
Abramson, J. et al. Accurate structure prediction of biomolecular interactions with AlphaFold 3. Nature 630, 493–500 (2024).
Balasubramaniam, M. et al. Aggregate interactome based on protein cross-linking interfaces predicts drug targets to limit aggregation in neurodegenerative diseases. iScience 20, 248–264 (2019).
Kim, D.-K. et al. A proteome-scale map of the SARS-CoV-2-human contactome. Nat. Biotechnol. 41, 140–149 (2023).
Mosca, R., Céol, A. & Aloy, P. Interactome3D: adding structural details to protein networks. Nat. Methods 10, 47–53 (2013).
Sambourg, L. & Thierry-Mieg, N. New insights into protein-protein interaction data lead to increased estimates of the S. cerevisiae interactome size. BMC Bioinform. 11, 605 (2010).
Yu, H. et al. High-quality binary protein interaction map of the yeast interactome network. Science 322, 104–110 (2008).
Stumpf, M. P. H. et al. Estimating the size of the human interactome. Proc. Natl. Acad. Sci. USA 105, 6959–6964 (2008).
Grigoriev, A. On the number of protein-protein interactions in the yeast proteome. Nucleic Acids Res. 31, 4157–4161 (2003).
Cusick, M. E. et al. Literature-curated protein interaction datasets. Nat. Methods 6, 39–46 (2009).
Luck, K. et al. A reference map of the human binary protein interactome. Nature 580, 402–408 (2020).
Cassonnet, P. et al. Benchmarking a luciferase complementation assay for detecting protein complexes. Nat. Methods 8, 990–992 (2011).
Eyckerman, S. et al. Design and application of a cytokine-receptor-based interaction trap. Nat. Cell Biol. 3, 1114–1119 (2001).
Uetz, P. et al. A comprehensive analysis of protein-protein interactions in Saccharomyces cerevisiae. Nature 403, 623–627 (2000).
Ito, T. et al. A comprehensive two-hybrid analysis to explore the yeast protein interactome. Proc. Natl. Acad. Sci. USA 98, 4569–4574 (2001).
Hu, Y. et al. Approaching a complete repository of sequence-verified protein-encoding clones for Saccharomyces cerevisiae. Genome Res. 17, 536–543 (2007).
Walhout, A. J. M. et al. Gateway recombinational cloning: application to the cloning of large numbers of open reading frames or ORFeomes. Methods Enzymol. 328, 575–592 (2000).
Choi, S. G. et al. Maximizing binary interactome mapping with a minimal number of assays. Nat. Commun. 10, 3907 (2019).
Costanzo, M. et al. A global genetic interaction network maps a wiring diagram of cellular function. Science 353, aaf1420 (2016).
Hillenmeyer, M. E. et al. The chemical genomic portrait of yeast: uncovering a phenotype for all genes. Science 320, 362–365 (2008).
Obayashi, T., Kagaya, Y., Aoki, Y., Tadaka, S. & Kinoshita, K. COXPRESdb v7: a gene coexpression database for 11 animal species supported by 23 coexpression platforms for technical evaluation and evolutionary inference. Nucleic Acids Res. 47, D55–D62 (2019).
Barabási, A.-L. & Oltvai, Z. N. Network biology: understanding the cell’s functional organization. Nat. Rev. Genet. 5, 101–113 (2004).
Huang, J. K. et al. Systematic evaluation of molecular networks for discovery of disease genes. Cell Syst. 6, 484–495.e5 (2018).
Baryshnikova, A. Systematic functional annotation and visualization of biological networks. Cell Syst. 2, 412–421 (2016).
Wood, V. et al. Hidden in plain sight: what remains to be discovered in the eukaryotic proteome? Open Biol. 9, 180241 (2019).
Venkatesan, K. et al. An empirical framework for binary interactome mapping. Nat. Methods 6, 83–90 (2009).
Kim, P. M., Sboner, A., Xia, Y. & Gerstein, M. The role of disorder in interaction networks: a structural analysis. Mol. Syst. Biol. 4, 179 (2008).
Siwek, J. C. et al. Sliding Window Interaction Grammar (SWING): a generalized interaction language model for peptide and protein interactions. Nat. Methods 22, 1707–1719 (2025).
Ming, Q., Gonzalez-Perez, D. & Luca, V. C. Molecular engineering strategies for visualizing low-affinity protein complexes. Exp. Biol. Med. 244, 1559–1567 (2019).
Matthews, L. R. et al. Identification of potential interaction networks using sequence-based searches for conserved protein-protein interactions or ‘interologs’. Genome Res. 11, 2120–2126 (2001).
Rual, J.-F. et al. Towards a proteome-scale map of the human protein–protein interaction network. Nature 437, 1173–1178 (2005).
Stelzl, U. et al. A human protein-protein interaction network: a resource for annotating the proteome. Cell 122, 957–968 (2005).
Rolland, T. et al. A proteome-scale map of the human interactome network. Cell 159, 1212–1226 (2014).
Haenig, C. et al. Interactome mapping provides a network of neurodegenerative disease proteins and uncovers widespread protein aggregation in affected brains. Cell Rep. 32, 108050 (2020).
Tompa, P., Davey, N. E., Gibson, T. J. & Babu, M. M. A million peptide motifs for the molecular biologist. Mol. Cell 55, 161–169 (2014).
Bryant, P., Pozzati, G. & Elofsson, A. Improved prediction of protein-protein interactions using AlphaFold2. Nat. Commun. 13, 1265 (2022).
Teufel, F. et al. Deorphanizing peptides using structure prediction. J. Chem. Inf. Model. 63, 2651–2655 (2023).
Burke, D. F. et al. Towards a structurally resolved human protein interaction network. Nat. Struct. Mol. Biol. 30, 216–225 (2023).
Kim, J.-A. & Haber, J. E. Chromatin assembly factors Asf1 and CAF-1 have overlapping roles in deactivating the DNA damage checkpoint when DNA repair is complete. Proc. Natl. Acad. Sci. USA 106, 1151–1156 (2009).
Tarasov, K. et al. High-content screening of yeast mutant libraries by shotgun lipidomics. Mol. Biosyst. 10, 1364–1376 (2014).
Huh, W.-K. et al. Global analysis of protein localization in budding yeast. Nature 425, 686–691 (2003).
Lee, M.-W. et al. Global protein expression profiling of budding yeast in response to DNA damage. Yeast 24, 145–154 (2007).
Bonetti, D., Anbalagan, S., Lucchini, G., Clerici, M. & Longhese, M. P. Tbf1 and Vid22 promote resection and non-homologous end joining of DNA double-strand break ends. EMBO J. 32, 275–289 (2013).
Marsh, J. A. & Teichmann, S. A. Structure, dynamics, assembly, and evolution of protein complexes. Annu. Rev. Biochem. 84, 551–575 (2015).
Aloy, P. & Russell, R. B. Ten thousand interactions for the molecular biologist. Nat. Biotechnol. 22, 1317–1321 (2004).
Díaz-Mejía, J. J. et al. Mapping DNA damage-dependent genetic interactions in yeast via party mating and barcode fusion genetics. Mol. Syst. Biol. 14, e7985 (2018).
Yachie, N. et al. Pooled-matrix protein interaction screens using barcode fusion genetics. Mol. Syst. Biol. 12, 863–863 (2016).
Mount, H. O. et al. A comprehensive two-hybrid analysis to explore the Legionella pneumophila effector-effector interactome. mSystems 9, e0100424 (2024).
Baryshev, A. et al. Massively parallel measurement of protein-protein interactions by sequencing using MP3-seq. Nat. Chem. Biol. 20, 1514–1523 (2024).
James, P., Halladay, J. & Craig, E. A. Genomic libraries and a host strain designed for highly efficient two-hybrid selection in yeast. Genetics 144, 1425–1436 (1996).
Dreze, M. et al. High-quality binary interactome mapping. Methods Enzymol. 470, 281–315 (2010).
Licata, L. et al. MINT, the molecular interaction database: 2012 update. Nucleic Acids Res. 40, D857–D861 (2012).
Orchard, S. et al. The MIntAct project–intact as a common curation platform for 11 molecular interaction databases. Nucleic Acids Res. 42, D358–D363 (2014).
Salwinski, L. et al. The database of interacting proteins: 2004 update. Nucleic Acids Res. 32, D449–D451 (2004).
Chatr-aryamontri, A. et al. The BioGRID interaction database: 2017 update. Nucleic Acids Res. 45, D369–D379 (2017).
Tarassov, K. et al. An in vivo map of the yeast protein interactome. Science 320, 1465–1470 (2008).
Gavin, A.-C. et al. Functional organization of the yeast proteome by systematic analysis of protein complexes. Nature 415, 141–147 (2002).
Gavin, A.-C. et al. Proteome survey reveals modularity of the yeast cell machinery. Nature 440, 631–636 (2006).
Krogan, N. J. et al. Global landscape of protein complexes in the yeast Saccharomyces cerevisiae. Nature 440, 637–643 (2006).
Ho, Y. et al. Systematic identification of protein complexes in Saccharomyces cerevisiae by mass spectrometry. Nature 415, 180–183 (2002).
Calderone, A., Castagnoli, L. & Cesareni, G. Mentha: a resource for browsing integrated protein-interaction networks. Nat. Methods 10, 690–691 (2013).
Bader, G. D., Betel, D. & Hogue, C. W. V. BIND: the biomolecular interaction network database. Nucleic Acids Res. 31, 248–250 (2003).
Kerrien, S. et al. The IntAct molecular interaction database in 2012. Nucleic Acids Res. 40, D841–D846 (2012).
Berman, H. M. et al. The protein data bank. Nucleic Acids Res. 28, 235–242 (2000).
Zhang, J. et al. Computing the human interactome. Preprint at bioRxiv https://doi.org/10.1101/2024.10.01.615885 (2024).
Hallez, R. et al. The asymmetric distribution of the essential histidine kinase PdhS indicates a differentiation event in Brucella abortus. EMBO J. 26, 1444–1455 (2007).
Vidalain, P.-O., Boxem, M., Ge, H., Li, S. & Vidal, M. Increasing specificity in high-throughput yeast two-hybrid experiments. Methods 32, 363–370 (2004).
Rual, J.-F. et al. Human ORFeome version 1.1: a platform for reverse proteomics. Genome Res. 14, 2128–2135 (2004).
Weile, J. et al. A framework for exhaustively mapping functional missense variants. Mol. Syst. Biol. 13, 957 (2017).
Tannous, B. A., Kim, D.-E., Fernandez, J. L., Weissleder, R. & Breakefield, X. O. Codon-optimized Gaussia luciferase cDNA for mammalian gene expression in culture and in vivo. Mol. Ther. 11, 435–443 (2005).
Mosca, R., Céol, A., Stein, A., Olivella, R. & Aloy, P. 3did: a catalog of domain-based interactions of known three-dimensional structure. Nucleic Acids Res. 42, D374–D379 (2014).
Dinkel, H. et al. ELM 2016–data update and new functionality of the eukaryotic linear motif resource. Nucleic Acids Res. 44, D294–D300 (2016).
Koh, J. L. Y. et al. CYCLoPs: a comprehensive database constructed from automated analysis of protein abundance and subcellular localization patterns in Saccharomyces cerevisiae. G3 Genes Genomes Genet. 5, 1223–1232 (2015).
Kanehisa, M., Sato, Y., Furumichi, M., Morishima, K. & Tanabe, M. New approach for understanding genome variations in KEGG. Nucleic Acids Res. 47, D590–D595 (2019).
Pu, S., Wong, J., Turner, B., Cho, E. & Wodak, S. J. Up-to-date catalogues of yeast protein complexes. Nucleic Acids Res. 37, 825–831 (2009).
Viger, F. & Latapy, M. Efficient and simple generation of random simple connected graphs with prescribed degree sequence. In Proc. 11th Annual International Computing and Combinatorics Conference, 440–449 (Springer Berlin, 2005).
Shannon, P. et al. Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome Res. 13, 2498–2504 (2003).
Barabási, A.-L. Network Science (Cambridge University Press, 2016).
Wu, Y., Li, R., Sun, S., Weile, J. & Roth, F. P. Improved pathogenicity prediction for rare human missense variants. Am. J. Hum. Genet. 108, 1891–1906 (2021).
Helmy, M. et al. OpenPIP: an open-source platform for hosting, visualizing and analyzing protein interaction data. J. Mol. Biol. 434, 167603 (2022).
Acknowledgements
We thank Steffi de Rouck for help with the MAPPIT experiments. We thank Gary Bader and acknowledge past and current members of the Center for Cancer Systems Biology (CCSB) for helpful discussions and experimental help. We thank Qian Cong and Jing Zhang for providing access to additional data files related to their AI PPI predictions. This work was funded by the following sources. National Institutes of Health grant R01HG006061 (M.V., D.E.H., M.A.C., M.E.C., and P.F.-B.). National Institutes of Health grant R01GM130885 (M.V.). National Institutes of Health grant R01GM133185 (M.V., M.A.C., and F.P.R.). Institute Sponsored Research funds from the Dana-Farber Cancer Institute Strategic Initiative (M.V.). Canadian Institutes of Health Research (CIHR) Foundation Grant FDN159926 (F.P.R.). Canadian Institutes of Health Research (CIHR) Project Grant PJT-162410 (J.R.). Léon Fredericq Foundation (A.D. and F.L.). Fund for Scientific Research (FRS-FNRS) Télévie Fellowships #7651317 F (A.D., J.-C.T.) and #7459421F (F.L., J.-C.T.). Natural Sciences and Engineering Research Council (NSERC) of Canada Banting Postdoctoral Fellowship (D.-K.K.). National Research Foundation (NRF) of Korea Basic Science Research Program grant 2017R1A6A3A03004385 funded by the Ministry of Education (D.-K.K.). National Institutes of Health National Resources For Network Biology (NRNB) Google Summer of Code 2015 (M.W.M.). U.S. National Science Foundation PHY-2440223 POLS NSF CAREER Award sponsored by NSF 22-586, and by the NSF–Simons National Institute for Theory and Mathematics in Biology, jointly funded by the U.S. National Science Foundation DMS-2235451 and the Simons Foundation MP-TMPS-00005320 (I.A.K). Spanish Ministry of Science Ramon y Cajal fellowship RYC-2017-22959 (C.P.). Dana-Farber Cancer Institute Center for Cancer Systems Biology (CCSB) Deborah F. Allinger Fellowships (A.Y., L.L.). Belgian American Educational Foundation (BAEF) Doctoral Research Fellowships (F.L.). Wallonia-Brussels International (WBI)-World Excellence Fellowships (F.L.). Herman-van Beneden Prize (F.L.). Josée and Jean Schmets Prize (F.L.). M.V. is a Chercheur Qualifié Honoraire, and J.-C.T. is a Directeur de Recherche from Fonds de la Recherche Scientifique (FRS-FNRS, Wallonia-Brussels Federation, Belgium). Free State of Bavaria’s AI for Therapy (AI4T) Initiative through the Institute of AI for Drug Discovery (AID) (P.F.-B.) and the Impuls and Networking Fund of the Helmholtz Association (PhenoPred) (P.F.-B.). J.D.L.R. acknowledges a Fulbright Grant Senior Scholar Grant (ref. PRX23/00628) awarded to work in the CCSB of the DFCI from January to July 2025.
Author information
Authors and Affiliations
Contributions
Computational analyses were performed by L.L., Y.W., with help from B.C., D.D.R., T.R., K.L., and O.D. Interactome mapping experiments were performed by A.D., T.C., with help from S.S., N.J., Q.Z., Z.Y., and K.S.-F. Sequencing to identify interacting proteins was carried out by A.G.C., M.G., N.K., J.J.K., and J.C.M. Y2H vectors were designed and generated by Q.Z. with help from N.J. The preparation of Y2H, GPCA, and MAPPIT destination clones by en masse gateway cloning and yeast transformations were performed by Q.Z., N.J., A.D., and T.C. Experimental results were processed by Y.W., T.H., and K.L. GPCA validation experiments were done by A.D., T.C., with help from Y.J. MAPPIT validation experiments were done by I.L., supervised by J.T. Y2H tests of predicted yeast interactions were performed by K.S.-F. Y2H tests of predicted human interactions were performed by F.L. and K.S.-F. with help from G.G.M. Functional enrichment analyses were done by D.-K.K., L.L., and Y.W. Extraction of the literature datasets was performed by L.L., T.H. YeRI web portal was built by M.W.M., supervised by J.R., M.H. Structural analyses were done by C.P., L.L., and Y.W., supervised by P.A. Images of 3D structural PPI models were produced by J.D.L.R. Topological analyses were done by L.L. Sequencing analyses were done by T.H., W.B., Y.S., and Y.W. Network-based functional prediction was performed by I.A.K. Additional experiments were performed by F.L., V.V.B.J., and G.M. The overall research effort was designed and conceptualized by M.V., F.P.R., M.A.C., D.E.H., P.F.-B., Y.W., A.D., L.L., and A.Y. Interactome mapping was supervised by B.C., M.V., M.A.C., D.E.H., and T.H. Manuscript was written and edited by L.L., A.Y., Y.W., A.D., T.H., F.L., F.P.R., J.D.L.R., P.F.-B., D.E.H., M.A.C., J.-C.T., and M.V. with contributions from other co-authors. The overall research effort was supervised and/or advised by M.V., F.P.R., M.A.C., and D.E.H. The project was conceived by M.V. Major funding acquisition was by M.V., D.E.H., M.A.C., F.P.R., P.F.-B., and M.E.C. D.-K.K., F.L., K.S.-F. contributed equally and should be considered co-second authors.
Corresponding authors
Ethics declarations
Competing interests
J.C.M. is a founder and CEO of seqWell, Inc; F.P.R., M.V. are shareholders and scientific advisors of seqWell, Inc. J.-C.T. is a founder of ExtraCell Biotech, SRL. The remaining authors declare no competing interests.
Peer review
Peer review information
Nature Communications thanks Ulrich Stelzl and the other anonymous reviewer(s) for their contribution to the peer review of this work. A peer review file is available.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
Source data
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Lambourne, L., Yadav, A., Wang, Y. et al. Experimental assessment of AI-based interactome mapping. Nat Commun (2026). https://doi.org/10.1038/s41467-026-70942-x
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41467-026-70942-x
