{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,27]],"date-time":"2026-07-27T14:15:23Z","timestamp":1785161723098,"version":"3.55.0"},"reference-count":47,"publisher":"American Chemical Society (ACS)","issue":"14","license":[{"start":{"date-parts":[[2026,7,13]],"date-time":"2026-07-13T00:00:00Z","timestamp":1783900800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,7,13]],"date-time":"2026-07-13T00:00:00Z","timestamp":1783900800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"},{"start":{"date-parts":[[2026,7,13]],"date-time":"2026-07-13T00:00:00Z","timestamp":1783900800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-045"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["12171434"],"award-info":[{"award-number":["12171434"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["62302165"],"award-info":[{"award-number":["62302165"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["U24A20249"],"award-info":[{"award-number":["U24A20249"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,7,27]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>Accurate identification of enzyme active sites is a prerequisite for elucidating protein functions and guiding enzyme engineering. Driven by the exponential growth of protein sequence data from next-generation sequencing, numerous novel enzymes with low sequence similarity have been discovered. While protein structure prediction models have provided highly accurate 3D structural data for these novel enzymes, the precise identification of their active sites remains a significant challenge for existing computational methods. Existing methods are constrained by static geometric representations, limited local receptive fields of graph encoders, and evolutionary semantic dilution. To address these limitations, this study presents Evo-EquiGPS, a multimodal graph neural network framework that synergizes multidimensional features for precise enzyme active site prediction. The model incorporates a three-branch parallel encoding architecture consisting of a dynamic geometric flow, a global topological flow, and an explicit evolutionary flow. It comprehensively integrates sequence semantics, 3D structures, and explicit evolutionary constraints of enzymes. Empirical evaluations demonstrate that Evo-EquiGPS exhibits superior performance across data sets with high structural diversity. On the TS124 data set, its area under the precision-recall curve (AUPRC) surpassed that of leading models such as SCREEN and GraphEC by a significant margin (exceeding SCREEN by 13.1% and GraphEC by 15.1%). Furthermore, the model demonstrated strong generalization capabilities on the highly diverse independent test set CSA112. Overall, the Evo-EquiGPS framework significantly enhances the precision of enzyme active site prediction. This provides a solid foundation for computational protein functional annotation.<\/jats:p>","DOI":"10.1021\/acs.jcim.6c00969","type":"journal-article","created":{"date-parts":[[2026,7,13]],"date-time":"2026-07-13T13:07:53Z","timestamp":1783948073000},"page":"8607-8619","source":"Crossref","is-referenced-by-count":0,"title":["Evo-EquiGPS: Synergizing\nDynamic Geometry, Global\nTopology, and Explicit Evolution for High-Precision Enzyme Active\nSite Prediction"],"prefix":"10.1021","volume":"66","author":[{"given":"Xinyu","family":"Fei","sequence":"first","affiliation":[{"name":"Huzhou Normal University , , , ,","place":["Huzhou, Zhejiang, China, 313000"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jiali","family":"Gu","sequence":"additional","affiliation":[{"name":"Huzhou Normal University , , , ,","place":["Huzhou, Zhejiang, China, 313000"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Cheng","family":"Zhou","sequence":"additional","affiliation":[{"name":"Huzhou Normal University , , , ,","place":["Huzhou, Zhejiang, China, 313000"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yue","family":"Liu","sequence":"additional","affiliation":[{"name":"Huzhou Normal University , , , ,","place":["Huzhou, Zhejiang, China, 313000"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zhong","family":"Li","sequence":"additional","affiliation":[{"name":"Huzhou Normal University , , , ,","place":["Huzhou, Zhejiang, China, 313000"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0004-6803-3551","authenticated-orcid":true,"given":"Yanlei","family":"Kang","sequence":"additional","affiliation":[{"name":"Huzhou Normal University , , , ,","place":["Huzhou, Zhejiang, China, 313000"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"316","published-online":{"date-parts":[[2026,7,13]]},"reference":[{"key":"2026072709090781800_cit1","doi-asserted-by":"publisher","DOI":"10.1016\/j.biotechadv.2024.108453","article-title":"Advances in multi-enzyme\nco-localization strategies for the construction of microbial cell\nfactory","volume":"77","author":"Guo","year":"2024","journal-title":"Biotechnol. Adv."},{"issue":"8016","key":"2026072709090781800_cit2","doi-asserted-by":"publisher","first-page":"493","DOI":"10.1038\/s41586-024-07487-w","article-title":"Accurate\nstructure prediction of biomolecular\ninteractions with AlphaFold 3","volume":"630","author":"Abramson","year":"2024","journal-title":"Nature"},{"issue":"7958","key":"2026072709090781800_cit3","doi-asserted-by":"publisher","first-page":"673","DOI":"10.1038\/s41586-023-05905-z","article-title":"Computational approaches streamlining drug discovery","volume":"616","author":"Sadybekov","year":"2023","journal-title":"Nature"},{"issue":"6","key":"2026072709090781800_cit4","doi-asserted-by":"publisher","first-page":"1587","DOI":"10.1007\/s10562-022-04137-6","article-title":"Redesigning Robust\nBiocatalysts by Engineering Enzyme\nMicroenvironment and Enzyme Immobilization","volume":"153","author":"Riaz","year":"2023","journal-title":"Catal. Lett."},{"issue":"1","key":"2026072709090781800_cit5","doi-asserted-by":"publisher","DOI":"10.1038\/s42003-024-06117-5","article-title":"A unified analysis\nof evolutionary and population constraint in protein domains highlights\nstructural features and pathogenic sites","volume":"7","author":"MacGowan","year":"2024","journal-title":"Commun.\nBiol."},{"issue":"2","key":"2026072709090781800_cit6","doi-asserted-by":"publisher","first-page":"45","DOI":"10.1080\/14789450.2025.2456046","article-title":"AlphaFold and what is next: bridging\nfunctional, systems\nand structural biology","volume":"22","author":"Szczepski","year":"2025","journal-title":"Expert Rev. Proteomics"},{"issue":"10","key":"2026072709090781800_cit7","doi-asserted-by":"publisher","DOI":"10.3390\/ijms24108846","article-title":"Enhanced Energetic\nPerformance via the Combination of Furoxan and Oxa-[5,5]bicyclic Structures","volume":"24","author":"Zhang","year":"2023","journal-title":"Int. J. Mol. Sci."},{"issue":"11","key":"2026072709090781800_cit8","doi-asserted-by":"publisher","first-page":"2308","DOI":"10.1002\/pro.5560061104","article-title":"TESS: a geometric hashing algorithm\nfor deriving 3D coordinate templates for searching structural databases.\nApplication to enzyme active sites","volume":"6","author":"Wallace","year":"1997","journal-title":"Protein\nSci."},{"issue":"6","key":"2026072709090781800_cit9","doi-asserted-by":"publisher","first-page":"864","DOI":"10.1093\/bioinformatics\/btu746","article-title":"GASS: identifying enzyme active sites\nwith genetic algorithms","volume":"31","author":"Izidoro","year":"2015","journal-title":"Bioinformatics"},{"issue":"14","key":"2026072709090781800_cit10","doi-asserted-by":"publisher","DOI":"10.1002\/pmic.201900119","article-title":"Protein Function Prediction: From Traditional Classifier\nto Deep Learning","volume":"19","author":"Lv","year":"2019","journal-title":"Proteomics"},{"issue":"1","key":"2026072709090781800_cit11","doi-asserted-by":"publisher","DOI":"10.1186\/s13059-019-1835-8","article-title":"The\nCAFA challenge reports\nimproved protein function prediction and new functional annotations\nfor hundreds of genes through experimental screens","volume":"20","author":"Zhou","year":"2019","journal-title":"Genome Biol."},{"issue":"20","key":"2026072709090781800_cit12","doi-asserted-by":"publisher","first-page":"2329","DOI":"10.1093\/bioinformatics\/btn433","article-title":"Accurate sequence-based prediction\nof catalytic residues","volume":"24","author":"Zhang","year":"2008","journal-title":"Bioinformatics"},{"key":"2026072709090781800_cit13","doi-asserted-by":"publisher","first-page":"125","DOI":"10.1016\/j.jtbi.2018.01.023","article-title":"PREvaIL, an integrative approach for inferring catalytic\nresidues using sequence, structural, and network features in a machine-learning\nframework","volume":"443","author":"Song","year":"2018","journal-title":"J. Theor. Biol."},{"key":"2026072709090781800_cit14","doi-asserted-by":"publisher","first-page":"157","DOI":"10.1016\/j.sbi.2020.01.005","article-title":"Using protein\nengineering to understand and modulate aggregation","volume":"60","author":"Ebo","year":"2020","journal-title":"Curr. Opin. Struct. Biol."},{"issue":"14","key":"2026072709090781800_cit15","doi-asserted-by":"publisher","first-page":"4277","DOI":"10.1021\/acs.jcim.3c00273","article-title":"A Highly Sensitive Model Based on\nGraph Neural Networks for Enzyme Key Catalytic Residue Prediction","volume":"63","author":"Shen","year":"2023","journal-title":"J. Chem. Inf. Model."},{"issue":"6","key":"2026072709090781800_cit16","doi-asserted-by":"publisher","DOI":"10.1093\/gpbjnl\/qzae094","article-title":"SCREEN: A Graph-based Contrastive\nLearning Tool to Infer Catalytic Residues and Assess Enzyme Mutations","volume":"22","author":"Pan","year":"2025","journal-title":"Genomics, Proteomics Bioinf."},{"key":"2026072709090781800_cit17","doi-asserted-by":"publisher","first-page":"1929","DOI":"10.1016\/j.csbj.2024.04.012","article-title":"Language models can identify enzymatic\nbinding sites in protein sequences","volume":"23","author":"Nana Teukam","year":"2024","journal-title":"Comput.\nStruct. Biotechnol. J."},{"issue":"6637","key":"2026072709090781800_cit18","doi-asserted-by":"publisher","first-page":"1123","DOI":"10.1126\/science.ade2574","article-title":"Evolutionary-scale prediction of atomic-level protein structure with\na language model","volume":"379","author":"Lin","year":"2023","journal-title":"Science"},{"issue":"15","key":"2026072709090781800_cit19","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.2016239118","article-title":"Biological structure\nand function emerge from scaling\nunsupervised learning to 250 million protein sequences","volume":"118","author":"Rives","year":"2021","journal-title":"Proc. Natl. Acad. Sci. U.S.A."},{"key":"2026072709090781800_cit20","doi-asserted-by":"crossref","unstructured":"Rao, R. M.; Liu, J.; Verkuil, R.; Meier, J.; Canny, J.; Abbeel, P.; Sercu, T.; Rives, A. In  MSA Transformer, Proceedings of the 38th International\nConference on Machine Learning; Proceedings of Machine\nLearning Research, 2021.","DOI":"10.1101\/2021.02.12.430858"},{"issue":"8","key":"2026072709090781800_cit21","doi-asserted-by":"publisher","first-page":"1099","DOI":"10.1038\/s41587-022-01618-2","article-title":"Large language models\ngenerate functional protein sequences across diverse families","volume":"41","author":"Madani","year":"2023","journal-title":"Nat. Biotechnol."},{"issue":"7873","key":"2026072709090781800_cit22","doi-asserted-by":"publisher","first-page":"583","DOI":"10.1038\/s41586-021-03819-2","article-title":"Highly accurate protein structure prediction\nwith AlphaFold","volume":"596","author":"Jumper","year":"2021","journal-title":"Nature"},{"issue":"9","key":"2026072709090781800_cit23","doi-asserted-by":"publisher","first-page":"704","DOI":"10.1038\/s41594-021-00650-1","article-title":"AlphaFold2\nand the future of structural biology","volume":"28","author":"Cramer","year":"2021","journal-title":"Nat. Struct.\nMol. Biol."},{"issue":"1","key":"2026072709090781800_cit24","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-024-52533-w","article-title":"Accurately predicting enzyme functions\nthrough geometric graph learning\non ESMFold-predicted structures","volume":"15","author":"Song","year":"2024","journal-title":"Nat. Commun."},{"issue":"1","key":"2026072709090781800_cit25","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-024-51511-6","article-title":"Multi-modal deep learning enables efficient and accurate annotation\nof enzymatic active sites","volume":"15","author":"Wang","year":"2024","journal-title":"Nat. Commun."},{"key":"2026072709090781800_cit26","doi-asserted-by":"publisher","DOI":"10.1016\/j.compbiomed.2025.110348","article-title":"A multimodal deep learning framework\nfor enzyme turnover prediction with missing modality","volume":"193","author":"Sun","year":"2025","journal-title":"Comput. Biol. Med."},{"issue":"12","key":"2026072709090781800_cit27","doi-asserted-by":"publisher","first-page":"1023","DOI":"10.1038\/s42256-021-00418-8","article-title":"Geometric\ndeep learning on molecular\nrepresentations","volume":"3","author":"Atz","year":"2021","journal-title":"Nat. Mach. Intelli."},{"key":"2026072709090781800_cit28","unstructured":"Dwivedi, V. P.; Bresson, X.\n          A Generalization\nof Transformer Networks to Graphs\n          ArXiv\n          2020, arXiv:2012.09699. arXiv.org e-Print archive. https:\/\/arxiv.org\/abs\/2012.09699."},{"key":"2026072709090781800_cit29","doi-asserted-by":"crossref","unstructured":"Li, G.; Mu\u0308ller, M.; Thabet, A. K.; Ghanem, B. In  DeepGCNs:\nCan GCNs\nGo As Deep As CNNs?, 2019 IEEE\/CVF International Conference\non Computer Vision (ICCV); IEEE, 2019; pp 9266\u20139275.","DOI":"10.1109\/ICCV.2019.00936"},{"issue":"45","key":"2026072709090781800_cit30","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.2406285121","article-title":"Protein language models learn evolutionary\nstatistics\nof interacting sequence motifs","volume":"121","author":"Zhang","year":"2024","journal-title":"Proc. Natl.\nAcad. Sci. U.S.A."},{"key":"2026072709090781800_cit31","unstructured":"Satorras, V. c. G.; Hoogeboom, E.; Welling, M. In  E(n) Equivariant Graph\nNeural Networks, Proceedings of the 38th International Conference\non Machine Learning; Proceedings of Machine Learning\nResearch, 2021."},{"issue":"2","key":"2026072709090781800_cit32","doi-asserted-by":"publisher","DOI":"10.1007\/s10462-025-11462-w","article-title":"Topological data analysis and topological deep\nlearning beyond persistent homology: a review","volume":"59","author":"Su","year":"2026","journal-title":"Artif. Intell. Rev."},{"key":"2026072709090781800_cit33","doi-asserted-by":"crossref","unstructured":"Dwivedi, V. P.; Rampa\u0301s\u030cek, L.; Galkin, M.; Parviz, A.; Wolf, G.; Luu, A. T.; Beaini, D.\n          Long Range Graph Benchmark. 2022, arXiv:2206.08164. arXiv.org e-Print archive. https:\/\/arxiv.org\/abs\/2206.08164.","DOI":"10.52202\/068431-1622"},{"key":"2026072709090781800_cit34","doi-asserted-by":"crossref","unstructured":"Rampa\u0301s\u030cek, L.; Galkin, M.; Dwivedi, V. P.; Luu, A. T.; Wolf, G.; Beaini, D.\n          Recipe\nfor a General, Powerful, Scalable Graph Transformer. 2022, arXiv:2205.12454. arXiv.org e-Print archive. https:\/\/arxiv.org\/abs\/2205.12454.","DOI":"10.52202\/068431-1054"},{"issue":"10","key":"2026072709090781800_cit35","doi-asserted-by":"publisher","first-page":"7112","DOI":"10.1109\/TPAMI.2021.3095381","article-title":"ProtTrans: Toward Understanding the Language\nof Life Through Self-Supervised Learning","volume":"44","author":"Elnaggar","year":"2022","journal-title":"IEEE\nTrans. Pattern Anal. Mach. Intel"},{"key":"2026072709090781800_cit36","doi-asserted-by":"crossref","unstructured":"Elnaggar, A.; Essam, H.; Salah-Eldin, W.; Moustafa, W.; Elkerdawy, M.; Rochereau, C.; Rost, B.\n          Ankh : Optimized Protein Language Model\nUnlocks General-Purpose Modelling.  bioRxiv2023, 10.1101\/2023.01.16.524265.","DOI":"10.1101\/2023.01.16.524265"},{"issue":"11","key":"2026072709090781800_cit37","doi-asserted-by":"publisher","first-page":"1026","DOI":"10.1038\/nbt.3988","article-title":"MMseqs2 enables sensitive protein\nsequence searching\nfor the analysis of massive data sets","volume":"35","author":"Steinegger","year":"2017","journal-title":"Nat.\nBiotechnol."},{"issue":"49","key":"2026072709090781800_cit38","doi-asserted-by":"publisher","first-page":"E1293","DOI":"10.1073\/pnas.1111471108","article-title":"Direct-coupling analysis of residue coevolution captures\nnative contacts across many protein families","volume":"108","author":"Morcos","year":"2011","journal-title":"Proc. Natl. Acad. Sci. U.S.A."},{"issue":"1","key":"2026072709090781800_cit39","doi-asserted-by":"publisher","first-page":"3942","DOI":"10.1609\/aaai.v32i1.11671","article-title":"FiLM: Visual\nReasoning with a General Conditioning Layer","volume":"32","author":"Perez","year":"2018","journal-title":"Proc. AAAI Conf. Artif. Intell."},{"issue":"4","key":"2026072709090781800_cit40","doi-asserted-by":"publisher","first-page":"719","DOI":"10.1016\/S0022-2836(03)00515-1","article-title":"Using A\nNeural Network and Spatial\nClustering to Predict the Location of Active Sites in Enzymes","volume":"330","author":"Gutteridge","year":"2003","journal-title":"J. Mol. Biol."},{"issue":"1","key":"2026072709090781800_cit41","doi-asserted-by":"publisher","DOI":"10.1186\/1471-2105-7-312","article-title":"Prediction of catalytic\nresidues using Support Vector Machine with\nselected protein sequence and structural properties","volume":"7","author":"Petrova","year":"2006","journal-title":"BMC Bioinf."},{"issue":"1","key":"2026072709090781800_cit42","doi-asserted-by":"publisher","DOI":"10.1186\/1471-2105-8-153","article-title":"How accurate and statistically robust are catalytic\nsite predictions based on closeness centrality?","volume":"8","author":"Chea","year":"2007","journal-title":"BMC Bioinf."},{"issue":"2","key":"2026072709090781800_cit43","doi-asserted-by":"publisher","first-page":"216","DOI":"10.1110\/ps.062523907","article-title":"Evaluation of features\nfor catalytic residue prediction in novel folds","volume":"16","author":"Youn","year":"2007","journal-title":"Protein Sci."},{"issue":"7","key":"2026072709090781800_cit100","doi-asserted-by":"publisher","first-page":"e41370","DOI":"10.1371\/journal.pone.0041370","article-title":"Identification\nof Catalytic Residues Using a Novel Feature that Integrates the Microenvironment\nand Geometrical Location Properties of Residues","volume":"7","author":"Han","year":"2012","journal-title":"PLOS ONE"},{"issue":"90001","key":"2026072709090781800_cit101","doi-asserted-by":"publisher","first-page":"D189","DOI":"10.1093\/nar\/gkh034","article-title":"The ASTRAL Compendium in 2004","volume":"32","author":"Chandonia","year":"2004","journal-title":"Nucleic Acids Research"},{"key":"2026072709090781800_cit44","doi-asserted-by":"publisher","DOI":"10.1016\/j.knosys.2026.115929","article-title":"Learning heterogeneous\nbiological interactions via meta-relation-guided dual-channel graph\ntransformer for circular ribonucleic acid function prediction","volume":"342","author":"Guo","year":"2026","journal-title":"Knowl.-Based Syst."},{"issue":"4","key":"2026072709090781800_cit45","doi-asserted-by":"publisher","DOI":"10.1007\/s11431-025-3216-y","article-title":"Nonfragile\nstate estimation for memristor based competitive neural networks subject\nto sensor saturation and stochastic cyber attacks","volume":"69","author":"Cao","year":"2026","journal-title":"Sci. China Technol. Sci."}],"container-title":["Journal of Chemical Information\nand Modeling"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/pubs.acs.org\/doi\/pdf\/10.1021\/acs.jcim.6c00969","content-type":"application\/pdf","content-version":"vor","intended-application":"unspecified"},{"URL":"https:\/\/pubs.acs.org\/jcisd8\/article-pdf\/66\/14\/8607\/65891218\/acs.jcim.6c00969.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/pubs.acs.org\/jcisd8\/article-pdf\/66\/14\/8607\/65891218\/acs.jcim.6c00969.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,27]],"date-time":"2026-07-27T13:33:04Z","timestamp":1785159184000},"score":1,"resource":{"primary":{"URL":"https:\/\/pubs.acs.org\/jcisd8\/article\/66\/14\/8607\/5207432\/Evo-EquiGPS-Synergizing-Dynamic-Geometry-Global"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,13]]},"references-count":47,"journal-issue":{"issue":"14","published-online":{"date-parts":[[2026,7,13]]},"published-print":{"date-parts":[[2026,7,27]]}},"URL":"https:\/\/doi.org\/10.1021\/acs.jcim.6c00969","relation":{},"ISSN":["1549-9596","1549-960X"],"issn-type":[{"value":"1549-9596","type":"print"},{"value":"1549-960X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,13]]}}}