{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,10]],"date-time":"2026-09-10T20:09:06Z","timestamp":1789070946292,"version":"build-2803163510"},"reference-count":52,"publisher":"Oxford University Press (OUP)","issue":"5","license":[{"start":{"date-parts":[[2026,9,10]],"date-time":"2026-09-10T00:00:00Z","timestamp":1788998400000},"content-version":"vor","delay-in-days":9,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,9,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>Metabolite\u2013protein interactions (MPIs) play pivotal roles in regulating cellular processes and metabolic pathways and have important implications for systems biology and drug discovery. Although several computational models have been proposed, alternative approaches are needed to broaden the methodological toolkit for MPI prediction. In this study, we developed a new deep-learning model for predicting MPIs. Chemical and protein large language models were used to generate informative representations of metabolites and proteins. For each entity type, we constructed two K-nearest-neighbor hypergraphs using different values of K, thereby capturing higher-order relationships at different scales and expanding the feature-learning space. A dual hypergraph convolutional network (HGCN) learned from these relationships. The resulting high-level features were further processed using a channel-wise attention mechanism and a one-dimensional convolutional neural network (1D-CNN) to obtain integrated metabolite and protein representations. Finally, a fully connected layer generated the predictions. The model was evaluated on four MPI datasets using global five-fold cross-validation; AUC and AUPR exceeded 0.9 on most datasets. Local five-fold cross-validation and an independent test further indicated that the model could predict MPIs involving previously unobserved metabolites or proteins. The model outperformed several existing MPI prediction models and related models adapted to this task. Ablation studies supported the rationale for its design.<\/jats:p>\n                  <jats:p>Short abstract In this study, we focused on the metabolite\u2013protein interactions (MPIs), which play pivotal roles in regulating cellular processes and metabolic pathways, and designed an effective deep-learning model for predicting MPIs. The model received the chemical and protein features yielded by large language models and refined them using a dual hypergraph convolutional network with two K-nearest-neighbor hypergraphs using different values of K. The resulting high-level features were further processed by a channel-wise attention mechanism and a one-dimensional convolutional neural network. A fully connected layer finally generated the predictions. The model shown competitive performance on four MPI datasets and outperformed several existing MPI prediction models.<\/jats:p>","DOI":"10.1093\/bib\/bbag489","type":"journal-article","created":{"date-parts":[[2026,8,19]],"date-time":"2026-08-19T11:47:05Z","timestamp":1787140025000},"source":"Crossref","is-referenced-by-count":0,"title":["PMPIHGLL: predicting metabolite\u2013protein interactions using dual hypergraph convolutional networks and large language models"],"prefix":"10.1093","volume":"27","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-3068-1583","authenticated-orcid":false,"given":"Lei","family":"Chen","sequence":"first","affiliation":[{"name":"College of Information Engineering, Shanghai Maritime University , No. 1550 Haigang Avenue, Pudong New Area, Shanghai 201306 ,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zhitong","family":"Jin","sequence":"additional","affiliation":[{"name":"College of Information Engineering, Shanghai Maritime University , No. 1550 Haigang Avenue, Pudong New Area, Shanghai 201306 ,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ying","family":"Shao","sequence":"additional","affiliation":[{"name":"College of Information Engineering, Shanghai Maritime University , No. 1550 Haigang Avenue, Pudong New Area, Shanghai 201306 ,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Bo","family":"Zhou","sequence":"additional","affiliation":[{"name":"School of Basic Medical Sciences, Shanghai University of Medicine and Health Sciences , No. 279 Zhouzhu Road, Pudong New Area, Shanghai 201318 ,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"286","published-online":{"date-parts":[[2026,9,10]]},"reference":[{"key":"2026091015403262100_ref1","doi-asserted-by":"publisher","first-page":"358","DOI":"10.1016\/j.cell.2017.12.006","article-title":"A map of protein-metabolite interactions reveals principles of chemical communication","volume":"172","author":"Piazza","year":"2018","journal-title":"Cell"},{"key":"2026091015403262100_ref2","doi-asserted-by":"publisher","first-page":"245","DOI":"10.1146\/annurev-biochem-061516-044757","article-title":"Systems biology of metabolism","volume":"86","author":"Nielsen","year":"2017","journal-title":"Annu Rev Biochem"},{"key":"2026091015403262100_ref3","doi-asserted-by":"publisher","first-page":"9","DOI":"10.1111\/j.1469-8137.2005.01513.x","article-title":"Regulation of metabolic networks: understanding metabolic complexity in the systems biology era","volume":"168","author":"Sweetlove","year":"2005","journal-title":"New Phytol"},{"key":"2026091015403262100_ref4","doi-asserted-by":"publisher","first-page":"747","DOI":"10.1038\/s41592-021-01197-1","article-title":"Mass spectrometry-based metabolomics: a guide for annotation, quantification and best reporting practices","volume":"18","author":"Alseekh","year":"2021","journal-title":"Nat Methods"},{"key":"2026091015403262100_ref5","doi-asserted-by":"publisher","first-page":"3774","DOI":"10.1021\/acs.jproteome.7b00457","article-title":"Accurate identification of unknown and known metabolic mixture components by combining 3D NMR with Fourier transform ion cyclotron resonance tandem mass spectrometry","volume":"16","author":"Wang","year":"2017","journal-title":"J Proteome Res"},{"key":"2026091015403262100_ref6","doi-asserted-by":"crossref","first-page":"vbac001","DOI":"10.1093\/bioadv\/vbac001","article-title":"Biclique extension as an effective approach to identify missing links in metabolic compound\u2013protein interaction networks","volume":"2","author":"Thieme","year":"2022","journal-title":"Bioinform Adv"},{"key":"2026091015403262100_ref7","doi-asserted-by":"crossref","first-page":"vbad098","DOI":"10.1093\/bioadv\/vbad098","article-title":"Prediction of metabolite-protein interactions based on integration of machine learning and constraint-based modeling","volume":"3","author":"Soleymani Babadi","year":"2023","journal-title":"Bioinform Adv"},{"key":"2026091015403262100_ref8","doi-asserted-by":"crossref","first-page":"lqae114","DOI":"10.1093\/nargab\/lqae114","article-title":"Machine learning of metabolite\u2013protein interactions from model-derived metabolic phenotypes","volume":"6","author":"Habibpour","year":"2024","journal-title":"NAR Genom Bioinform"},{"key":"2026091015403262100_ref9","doi-asserted-by":"crossref","first-page":"bbab014","DOI":"10.1093\/bib\/bbab014","article-title":"Prediction and collection of protein-metabolite interactions","volume":"22","author":"Zhao","year":"2021","journal-title":"Brief Bioinform"},{"key":"2026091015403262100_ref10","doi-asserted-by":"crossref","first-page":"bbad189","DOI":"10.1093\/bib\/bbad189","article-title":"MPI-VGAE: protein\u2013metabolite enzymatic reaction link learning by variational graph autoencoders","volume":"24","author":"Wang","year":"2023","journal-title":"Brief Bioinform"},{"key":"2026091015403262100_ref11","doi-asserted-by":"publisher","first-page":"1297","DOI":"10.1038\/s42256-023-00740-3","article-title":"Neural scaling of deep chemical models","volume":"5","author":"Frey","year":"2023","journal-title":"Nature Machine Intelligence"},{"key":"2026091015403262100_ref12","doi-asserted-by":"publisher","first-page":"7112","DOI":"10.1109\/TPAMI.2021.3095381","article-title":"Prottrans: toward understanding the language of life through self-supervised learning","volume":"44","author":"Elnaggar","year":"2022","journal-title":"IEEE Trans Pattern Anal Mach Intell"},{"key":"2026091015403262100_ref13","doi-asserted-by":"publisher","first-page":"D380","DOI":"10.1093\/nar\/gkv1277","article-title":"STITCH 5: augmenting protein-chemical interaction networks with tissue and affinity data","volume":"44","author":"Szklarczyk","year":"2016","journal-title":"Nucleic Acids Res"},{"key":"2026091015403262100_ref14","doi-asserted-by":"publisher","first-page":"D638","DOI":"10.1093\/nar\/gkac1000","article-title":"The STRING database in 2023: protein-protein association networks and functional enrichment analyses for any sequenced genome of interest","volume":"51","author":"Szklarczyk","year":"2023","journal-title":"Nucleic Acids Res"},{"key":"2026091015403262100_ref15","doi-asserted-by":"crossref","first-page":"bbae386","DOI":"10.1093\/bib\/bbae386","article-title":"PMiSLocMF: predicting miRNA subcellular localizations by incorporating multi-source features of miRNAs","volume":"25","author":"Chen","year":"2024","journal-title":"Brief Bioinform"},{"key":"2026091015403262100_ref16","doi-asserted-by":"crossref","first-page":"bbac379","DOI":"10.1093\/bib\/bbac379","article-title":"GraphCDA: a hybrid graph representation learning framework based on GCN and GAT for predicting disease-associated circRNAs","volume":"23","author":"Dai","year":"2022","journal-title":"Brief Bioinform"},{"key":"2026091015403262100_ref17","doi-asserted-by":"publisher","first-page":"105322","DOI":"10.1016\/j.compbiomed.2022.105322","article-title":"RGCNCDA: relational graph convolutional network improves circRNA-disease association prediction by incorporating microRNAs","volume":"143","author":"Chen","year":"2022","journal-title":"Comput Biol Med"},{"key":"2026091015403262100_ref18","doi-asserted-by":"publisher","first-page":"1765","DOI":"10.1109\/TCBB.2024.3415058","article-title":"SAGCN: using graph convolutional network with subgraph-aware for circRNA-drug sensitivity identification","volume":"21","author":"Sun","year":"2024","journal-title":"IEEE\/ACM Transactions on Computational Biology and Bioinformatics"},{"key":"2026091015403262100_ref19","doi-asserted-by":"crossref","first-page":"bbaa243","DOI":"10.1093\/bib\/bbaa243","article-title":"Predicting drug-disease associations through layer attention graph convolutional network","volume":"22","author":"Yu","year":"2021","journal-title":"Brief Bioinform"},{"key":"2026091015403262100_ref20","doi-asserted-by":"publisher","first-page":"107637","DOI":"10.1016\/j.patcog.2020.107637","article-title":"Hypergraph convolution and hypergraph attention","volume":"110","author":"Bai","year":"2021","journal-title":"Pattern Recogn"},{"key":"2026091015403262100_ref21","doi-asserted-by":"publisher","first-page":"230","DOI":"10.1038\/s41598-024-81862-5","article-title":"Predicting noncoding RNA and disease associations using multigraph contrastive learning","volume":"15","author":"Sun","year":"2025","journal-title":"Sci Rep"},{"key":"2026091015403262100_ref22","doi-asserted-by":"crossref","first-page":"291","DOI":"10.1186\/s12911-023-02384-0","article-title":"MNCLCDA: predicting circRNA-drug sensitivity associations by using mixed neighbourhood information and contrastive learning","volume":"23","author":"Li","year":"2023","journal-title":"BMC Med Inform Decis Mak"},{"key":"2026091015403262100_ref23","doi-asserted-by":"publisher","first-page":"2847","DOI":"10.1093\/bioinformatics\/btac164","article-title":"Supervised graph co-contrastive learning for drug\u2013target interaction prediction","volume":"38","author":"Li","year":"2022","journal-title":"Bioinformatics"},{"key":"2026091015403262100_ref24","doi-asserted-by":"publisher","first-page":"12099","DOI":"10.1021\/acs.jcim.5c01968","article-title":"HCLAMCMI: prediction of circRNA-miRNA interactions based on hypergraph contrastive learning and an attention mechanism","volume":"65","author":"Chen","year":"2025","journal-title":"J Chem Inf Model"},{"key":"2026091015403262100_ref25","volume-title":"3rd International Conference on Learning Representations","author":"Kingma","year":"2015"},{"key":"2026091015403262100_ref26","first-page":"1137","volume-title":"International Joint Conference on Artificial Intelligence","author":"Kohavi","year":"1995"},{"key":"2026091015403262100_ref27","doi-asserted-by":"publisher","DOI":"10.2174\/0115748936408819251114062606","article-title":"Identification of gene signatures associated with multisystem inflammatory syndrome in children after SARS-CoV-2 infection","author":"Ren","year":"2026","journal-title":"Curr Bioinforma"},{"key":"2026091015403262100_ref28","doi-asserted-by":"publisher","first-page":"13","DOI":"10.1016\/j.cancergen.2026.01.003","article-title":"Identifying transcriptional signatures of leukocytes in tissue and blood for multicancer diagnosis by using machine learning methods","volume":"302-303","author":"Ma","year":"2026","journal-title":"Cancer Genetics"},{"key":"2026091015403262100_ref29","doi-asserted-by":"publisher","first-page":"141113","DOI":"10.1016\/j.bbapap.2025.141113","article-title":"Integrative multi-omics machine learning reveals novel driver genes associations in lung adenocarcinoma, Biochimica et Biophysica Acta (BBA)","volume":"1874","author":"Yuan","year":"2026","journal-title":"Proteins and Proteomics"},{"key":"2026091015403262100_ref30","doi-asserted-by":"publisher","first-page":"4256","DOI":"10.1093\/bioinformatics\/bty503","article-title":"Predicting miRNA-disease association based on inductive matrix completion","volume":"34","author":"Chen","year":"2018","journal-title":"Bioinformatics"},{"key":"2026091015403262100_ref31","first-page":"37","article-title":"Evaluation: from precision, recall and f-measure to roc., informedness, markedness & correlation","volume":"2","author":"Powers","year":"2011","journal-title":"J Mach Learn Technol"},{"key":"2026091015403262100_ref32","doi-asserted-by":"publisher","DOI":"10.2174\/0115748936434191260212112236","article-title":"AntiCanNet: a graph convolution and chemical LLM framework for predicting anti-cancer small molecules","author":"Chen","year":"2026","journal-title":"Curr Bioinforma"},{"key":"2026091015403262100_ref33","doi-asserted-by":"crossref","first-page":"32","DOI":"10.1186\/s12859-025-06356-1","article-title":"PLysPTM-HGNN: predicting lysine PTM sites of proteins using hybrid graph neural networks","volume":"27","author":"Chen","year":"2026","journal-title":"BMC Bioinformatics"},{"key":"2026091015403262100_ref34","doi-asserted-by":"crossref","first-page":"4876","DOI":"10.1038\/s41598-026-35110-7","article-title":"Root-associated protein prediction using a protein large language model and hypergraph convolutional networks","volume":"16","author":"Chen","year":"2026","journal-title":"Sci Rep"},{"key":"2026091015403262100_ref35","doi-asserted-by":"publisher","first-page":"5061","DOI":"10.1093\/bioinformatics\/btaa577","article-title":"MDIPA: a microRNA-drug interaction prediction approach based on non-negative matrix factorization","volume":"36","author":"Jamali","year":"2020","journal-title":"Bioinformatics"},{"key":"2026091015403262100_ref36","doi-asserted-by":"crossref","first-page":"bbac104","DOI":"10.1093\/bib\/bbac104","article-title":"Identification of miRNA-disease associations via deep forest ensemble learning based on autoencoder","volume":"23","author":"Liu","year":"2022","journal-title":"Brief Bioinform"},{"key":"2026091015403262100_ref37","doi-asserted-by":"crossref","first-page":"12775","DOI":"10.1038\/s41598-026-43808-x","article-title":"Predicting circRNA subcellular localization by fusing circRNA sequence and network information","volume":"16","author":"Chen","year":"2026","journal-title":"Sci Rep"},{"key":"2026091015403262100_ref38","doi-asserted-by":"publisher","first-page":"742","DOI":"10.1021\/ci100050t","article-title":"Extended-connectivity fingerprints","volume":"50","author":"Rogers","year":"2010","journal-title":"J Chem Inf Model"},{"key":"2026091015403262100_ref39","doi-asserted-by":"publisher","first-page":"1466","DOI":"10.1002\/jcc.21707","article-title":"PaDEL-descriptor: an open source software to calculate molecular descriptors and fingerprints","volume":"32","author":"Yap","year":"2011","journal-title":"J Comput Chem"},{"key":"2026091015403262100_ref40","doi-asserted-by":"publisher","first-page":"4337","DOI":"10.1073\/pnas.0607879104","article-title":"Predicting protein\u2013protein interactions based only on sequences information","volume":"104","author":"Shen","year":"2007","journal-title":"Proc Natl Acad Sci"},{"key":"2026091015403262100_ref41","doi-asserted-by":"publisher","first-page":"10915","DOI":"10.1073\/pnas.89.22.10915","article-title":"Amino acid substitution matrices from protein blocks","volume":"89","author":"Henikoff","year":"1992","journal-title":"Proc Natl Acad Sci USA"},{"key":"2026091015403262100_ref42","doi-asserted-by":"crossref","first-page":"308","DOI":"10.1109\/TCBB.2010.93","article-title":"On position-specific scoring matrix for protein function prediction","volume":"8","author":"Cheol Jeong","year":"2010","journal-title":"IEEE\/ACM Transactions on Computational Biology and Bioinformatics"},{"key":"2026091015403262100_ref43","doi-asserted-by":"crossref","first-page":"14938","DOI":"10.1038\/s41598-017-14945-1","article-title":"iDNAProt-ES: identification of DNA-binding proteins using evolutionary and structural features","volume":"7","author":"Chowdhury","year":"2017","journal-title":"Sci Rep"},{"key":"2026091015403262100_ref44","doi-asserted-by":"publisher","first-page":"1123","DOI":"10.1126\/science.ade2574","article-title":"Evolutionary-scale prediction of atomic-level protein structure with a language model","volume":"379","author":"Lin","year":"2023","journal-title":"Science"},{"key":"2026091015403262100_ref45","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1186\/s40649-019-0069-y","article-title":"Graph convolutional networks: a comprehensive review","volume":"6","author":"Zhang","year":"2019","journal-title":"Comput Soc Netw"},{"key":"2026091015403262100_ref46","volume-title":"Graph Attention Networks","author":"Veli\u010dkovi\u0107","year":"2018"},{"key":"2026091015403262100_ref47","first-page":"1024","volume-title":"Inductive Representation Learning on Large Graphs","author":"Hamilton","year":"2017"},{"key":"2026091015403262100_ref48","first-page":"12278 2020","article-title":"Hnhn: hypergraph networks with hyperedge neurons","author":"Dong","year":"2006","journal-title":"arXiv preprint arXiv"},{"key":"2026091015403262100_ref49","article-title":"Be more with less: hypergraph attention networks for inductive text classification","author":"Ding","year":"2020","journal-title":"arXiv preprint arXiv:201100387"},{"key":"2026091015403262100_ref50","article-title":"Hypersage: generalizing inductive representation learning on hypergraphs","author":"Arya","year":"2020","journal-title":"arXiv preprint arXiv:201004558"},{"key":"2026091015403262100_ref51","doi-asserted-by":"publisher","first-page":"171","DOI":"10.1016\/j.ymben.2023.06.001","article-title":"Rank-ordering of known enzymes as starting points for re-engineering novel substrate activity using a convolutional neural network","volume":"78","author":"Upadhyay","year":"2023","journal-title":"Metab Eng"},{"key":"2026091015403262100_ref52","doi-asserted-by":"publisher","first-page":"187","DOI":"10.1021\/acs.jcim.4c01801","article-title":"Deep learning-driven insights into enzyme-substrate interaction discovery","volume":"65","author":"Qian","year":"2025","journal-title":"J Chem Inf Model"}],"container-title":["Briefings in Bioinformatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/bib\/article-pdf\/27\/5\/bbag489\/71022378\/bbag489.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bib\/article-pdf\/27\/5\/bbag489\/71022378\/bbag489.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,10]],"date-time":"2026-09-10T19:40:40Z","timestamp":1789069240000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/bib\/article\/doi\/10.1093\/bib\/bbag489\/8790180"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,1]]},"references-count":52,"journal-issue":{"issue":"5","published-print":{"date-parts":[[2026,9,1]]}},"URL":"https:\/\/doi.org\/10.1093\/bib\/bbag489","relation":{},"ISSN":["1467-5463","1477-4054"],"issn-type":[{"value":"1467-5463","type":"print"},{"value":"1477-4054","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2026,9]]},"published":{"date-parts":[[2026,9,1]]},"article-number":"bbag489"}}