{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,3]],"date-time":"2026-08-03T05:40:07Z","timestamp":1785735607475,"version":"3.56.0"},"reference-count":32,"publisher":"Oxford University Press (OUP)","issue":"7","license":[{"start":{"date-parts":[[2026,6,15]],"date-time":"2026-06-15T00:00:00Z","timestamp":1781481600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"AGAUR research","award":["2021SGR-00111"],"award-info":[{"award-number":["2021SGR-00111"]}]},{"name":"ASCLEPIUS: Smart Technology for Smart Healthcare","award":["PID2022-138327OB-I00"],"award-info":[{"award-number":["PID2022-138327OB-I00"]}]},{"name":"Ministerio de Ciencia e Innovaci\u00f3n (MCIN)\/Agencia Estatal de Investigaci\u00f3n","award":["(AEI)\/10.13039\/501100011033\/FEDER"],"award-info":[{"award-number":["(AEI)\/10.13039\/501100011033\/FEDER"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,7,2]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Motivation<\/jats:title>\n                    <jats:p>Graphs have been demonstrated to have an impressive ability to keep the structural and semantic properties of chemical compounds and are therefore widely used in molecular modeling. Moreover, Graph Neural Networks (GNNs) and Large Language Models (LLMs) have achieved outstanding results in predicting molecular properties and generating textual descriptions given graphs or texts, respectively. Recently, some mathematical models have been presented for molecular science applications, which combine the GNNs ability to capture the structural and semantic information and the generative ability of LLMs. However, these models are dispersed across the literature and vary in architecture, input\u2013output design, and application scope, making it difficult to systematically compare them or to identify suitable approaches for specific research objectives.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>This paper classifies recent GNN\u2013LLM models and summarizes them with the aim of providing guidance for research involving their use or the development of new models. We present tables that depict the properties and parameters of the models, as well as their associated chemical computational applications. In addition, a new model classification is presented to help researchers define the models they use or future ones. Finally, we quantitatively compare these models based on their reported experimental results.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1093\/bioinformatics\/btag387","type":"journal-article","created":{"date-parts":[[2026,6,15]],"date-time":"2026-06-15T19:04:38Z","timestamp":1781550278000},"source":"Crossref","is-referenced-by-count":1,"title":["A survey of models composed of graph neural networks and large language models for molecular science"],"prefix":"10.1093","volume":"42","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-6276-2049","authenticated-orcid":false,"given":"Nat\u00e0lia","family":"Segura-Alabart","sequence":"first","affiliation":[{"name":"Departament d\u2019Enginyeria Inform\u00e0tica i Matem\u00e0tiques, Universitat Rovira i Virgili , Tarragona, Catalonia 43007,","place":["Spain"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6112-5913","authenticated-orcid":false,"given":"Francesc","family":"Serratosa","sequence":"additional","affiliation":[{"name":"Departament d\u2019Enginyeria Inform\u00e0tica i Matem\u00e0tiques, Universitat Rovira i Virgili , Tarragona, Catalonia 43007,","place":["Spain"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"286","published-online":{"date-parts":[[2026,6,15]]},"reference":[{"key":"2026070212465753000_btag387-B1","article-title":"Prot2text: multimodal protein\u2019s function generation with gnns and transformers","author":"Abdine","year":"2023","journal-title":"CoRR"},{"key":"2026070212465753000_btag387-B2","first-page":"354","author":"Cao","year":"2025"},{"key":"2026070212465753000_btag387-B3","first-page":"6140","author":"Christofidellis","year":"2023"},{"key":"2026070212465753000_btag387-B4","doi-asserted-by":"publisher","first-page":"5237","DOI":"10.1021\/acs.jctc.6c00026","article-title":"MolXProt: a cross-attention transformer-based graph neural network for protein-ligand binding affinity prediction","volume":"22","author":"Cucco","year":"2026","journal-title":"J Chem Theory Comput"},{"key":"2026070212465753000_btag387-B5","first-page":"595","author":"Edwards","year":"2021"},{"key":"2026070212465753000_btag387-B6","doi-asserted-by":"crossref","first-page":"1582","DOI":"10.2174\/1568026620666200603122000","article-title":"Learning the edit costs of graph edit distance applied to ligand-based virtual screening","volume":"20","author":"Garcia-Hernandez","year":"2020","journal-title":"Curr Top Med Chem"},{"key":"2026070212465753000_btag387-B7","doi-asserted-by":"crossref","first-page":"8622","DOI":"10.1109\/TKDE.2024.3469578","article-title":"Large language models on graphs: a comprehensive survey","volume":"36","author":"Jin","year":"2024","journal-title":"IEEE Trans Knowl Data Eng"},{"key":"2026070212465753000_btag387-B8","author":"Kipf","year":"2016"},{"key":"2026070212465753000_btag387-B9","doi-asserted-by":"crossref","first-page":"045024","DOI":"10.1088\/2632-2153\/aba947","article-title":"Self-referencing embedded strings (SELFIES): a 100% robust molecular string representation","volume":"1","author":"Krenn","year":"2020","journal-title":"Mach Learn Sci Technol"},{"key":"2026070212465753000_btag387-B10","first-page":"19730","author":"Li","year":"2023"},{"key":"2026070212465753000_btag387-B11","doi-asserted-by":"publisher","first-page":"82","DOI":"10.1186\/s12859-026-06407-1","article-title":"Multimodal learning on heterogeneous subgraphs and LLMs representation for MHC-peptide binding affinity prediction","volume":"27","author":"Li","year":"2026","journal-title":"BMC Bioinformatics"},{"key":"2026070212465753000_btag387-B12","first-page":"8123","author":"Li","year":"2024"},{"key":"2026070212465753000_btag387-B13","doi-asserted-by":"publisher","author":"Lin","year":"2022","DOI":"10.1101\/2022.07.20.500902"},{"key":"2026070212465753000_btag387-B14","author":"Liu","year":"2025"},{"key":"2026070212465753000_btag387-B15","doi-asserted-by":"crossref","first-page":"108073","DOI":"10.1016\/j.compbiomed.2024.108073","article-title":"GIT-Mol: a multi-modal large language model for molecular science with graph, image, and text","volume":"171","author":"Liu","year":"2024","journal-title":"Comput Biol Med"},{"key":"2026070212465753000_btag387-B16","doi-asserted-by":"crossref","first-page":"1447","DOI":"10.1038\/s42256-023-00759-6","article-title":"Multi-modal molecule structure\u2013text model for text-based retrieval and editing","volume":"5","author":"Liu","year":"2023","journal-title":"Nat Mach Intell"},{"key":"2026070212465753000_btag387-B17","first-page":"15623","author":"Liu","year":"2023"},{"key":"2026070212465753000_btag387-B18","first-page":"1606","author":"Liu","year":"2023"},{"key":"2026070212465753000_btag387-B19","doi-asserted-by":"crossref","first-page":"80","DOI":"10.1016\/j.patrec.2025.04.032","article-title":"AE+GAE: a mixture model for structural and semantic feature detection in graph neural network regression","volume":"195","author":"Serratosa","year":"2025","journal-title":"Pattern Recognit Lett"},{"key":"2026070212465753000_btag387-B20","author":"Shi","year":"2023"},{"key":"2026070212465753000_btag387-B21","author":"Su","year":"2022"},{"key":"2026070212465753000_btag387-B22","doi-asserted-by":"publisher","DOI":"10.48550\/ARXIV.2505.22948","article-title":"Foundation molecular grammar: multi-modal foundation models induce interpretable molecular graph languages","volume-title":"CoRR","author":"Sun","year":"2025"},{"key":"2026070212465753000_btag387-B23","doi-asserted-by":"crossref","first-page":"6295","DOI":"10.1007\/s10462-022-10321-2","article-title":"A survey of graph neural networks in various learning paradigms: methods, applications, and challenges","volume":"56","author":"Waikhom","year":"2023","journal-title":"Artif Intell Rev"},{"key":"2026070212465753000_btag387-B24","doi-asserted-by":"crossref","first-page":"108215","DOI":"10.1016\/j.patcog.2021.108215","article-title":"Graph convolutional autoencoders with co-learning of graph structure and node attributes","volume":"121","author":"Wang","year":"2022","journal-title":"Pattern Recognit"},{"key":"2026070212465753000_btag387-B25","doi-asserted-by":"publisher","first-page":"116","DOI":"10.1186\/s12859-025-06123-2","article-title":"PPI-Graphomer: enhanced protein-protein affinity prediction using pretrained and graph transformer models","volume":"26","author":"Xie","year":"2025","journal-title":"BMC Bioinformatics"},{"key":"2026070212465753000_btag387-B26","author":"Xu","year":"2019"},{"key":"2026070212465753000_btag387-B27","first-page":"28877","author":"Ying","year":"2021"},{"key":"2026070212465753000_btag387-B28","doi-asserted-by":"crossref","first-page":"104","DOI":"10.1016\/j.neunet.2022.05.026","article-title":"Graph transformer networks: learning meta-path graphs to improve GNNs","volume":"153","author":"Yun","year":"2022","journal-title":"Neural Netw"},{"key":"2026070212465753000_btag387-B29","doi-asserted-by":"crossref","first-page":"690049","DOI":"10.3389\/fgene.2021.690049","article-title":"Graph neural networks and their current applications in bioinformatics","volume":"12","author":"Zhang","year":"2021","journal-title":"Front Genet"},{"key":"2026070212465753000_btag387-B30","first-page":"1","article-title":"Explainability for large language models: a survey","volume":"15","author":"Zhao","year":"2024","journal-title":"ACM Trans Intell Syst Technol"},{"key":"2026070212465753000_btag387-B31","article-title":"GIMLET: a unified graph-text model for instruction-based molecule zero-shot learning","volume":"36","author":"Zhao","year":"2023","journal-title":"Adv Neural Inf Process Syst"},{"key":"2026070212465753000_btag387-B32","doi-asserted-by":"crossref","first-page":"57","DOI":"10.1016\/j.aiopen.2021.01.001","article-title":"Graph neural networks: a review of methods and applications","volume":"1","author":"Zhou","year":"2020","journal-title":"AI Open"}],"container-title":["Bioinformatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/bioinformatics\/advance-article-pdf\/doi\/10.1093\/bioinformatics\/btag387\/68530133\/btag387.pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/42\/7\/btag387\/68530133\/btag387.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/42\/7\/btag387\/68530133\/btag387.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,2]],"date-time":"2026-07-02T17:00:20Z","timestamp":1783011620000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article\/doi\/10.1093\/bioinformatics\/btag387\/8708323"}},"subtitle":[],"editor":[{"given":"Arne","family":"Elofsson","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2026,6,15]]},"references-count":32,"journal-issue":{"issue":"7","published-print":{"date-parts":[[2026,7,2]]}},"URL":"https:\/\/doi.org\/10.1093\/bioinformatics\/btag387","relation":{},"ISSN":["1367-4803","1367-4811"],"issn-type":[{"value":"1367-4803","type":"print"},{"value":"1367-4811","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2026,7]]},"published":{"date-parts":[[2026,6,15]]},"article-number":"btag387"}}