{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,15]],"date-time":"2026-09-15T15:42:18Z","timestamp":1789486938786,"version":"build-2803163510"},"reference-count":47,"publisher":"Oxford University Press (OUP)","issue":"9","license":[{"start":{"date-parts":[[2026,8,27]],"date-time":"2026-08-27T00:00:00Z","timestamp":1787788800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Sanofi Inc"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,8,31]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Timeseries clinical transcriptomic datasets offer the opportunity to gain insights into the dynamics of disease mechanisms\/treatment responses. However, their utility in uncovering temporal patterns is often limited by high noise levels and small sample sizes. Leveraging foundational gene embeddings and incorporating interaction information can help address these challenges, improve gene network analysis, and enable the detection of subtle changes that drive disease progression or drug response.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>We finetuned gene embeddings from foundation models using healthy tissue gene expression data and used them in temporal GNNs to model gene expression of responder and non-responders to treatment in four disease datasets\u2014ulcerative colitis, Crohn\u2019s disease, Alopecia Areata, and psoriasis. Application of our method to these datasets confirmed known mechanisms associated with drug action, and also identified key differences between activated and repressed pathways for responders and non-responders, including B-Cell activation and mitochondria-related activity in ulcerative colitis patients.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>Finetuning gene embeddings from foundation models provides a richer context to model gene expression data compared to using them in their naive state. Even with smaller sample sizes, results from GNN-based temporal models outperform traditional methods by detecting known mechanisms of response and unraveling role of genes and mechanisms not known to be associated with response and non-response.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Code availability<\/jats:title>\n                    <jats:p>Code and data are available in a public GitHub repository\u2014https:\/\/github.com\/Sanofi-Public\/GNN-Timeseries. DOI: 10.5281\/zenodo.20494035.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1093\/bioinformatics\/btag640","type":"journal-article","created":{"date-parts":[[2026,8,27]],"date-time":"2026-08-27T13:26:12Z","timestamp":1787837172000},"source":"Crossref","is-referenced-by-count":0,"title":["Finetuning Foundation Models for Temporal Clinical Transcriptomics Data"],"prefix":"10.1093","volume":"42","author":[{"ORCID":"https:\/\/orcid.org\/0009-0008-9899-0458","authenticated-orcid":false,"given":"Sachin","family":"Mathur","sequence":"first","affiliation":[{"name":"R&D Data and Computational Sciences, Sanofi , Cambridge, MA 02141,","place":["United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Alexander","family":"Kagan","sequence":"additional","affiliation":[{"name":"Department of Statistics, University of Michigan , Ann Arbor, MI 48109,","place":["United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Peyman","family":"Passban","sequence":"additional","affiliation":[{"name":"R&D Data and Computational Sciences, Sanofi , Toronto, Ontario M5V 1V6,","place":["Canada"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hamid","family":"Mattoo","sequence":"additional","affiliation":[{"name":"Targets Disease Systems Biology, Sanofi , Cambridge, MA 02141,","place":["United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Euxhen","family":"Hasanaj","sequence":"additional","affiliation":[{"name":"GenBio AI , Palo Alto, CA 94301,","place":["United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ziv","family":"Bar-Joseph","sequence":"additional","affiliation":[{"name":"GenBio AI , Palo Alto, CA 94301,","place":["United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"286","published-online":{"date-parts":[[2026,8,27]]},"reference":[{"key":"2026091509361866500_btag640-B1","doi-asserted-by":"crossref","first-page":"1307394","DOI":"10.3389\/fmed.2024.1307394","article-title":"Exploring the role of il-1\u03b2 in inflammatory bowel disease pathogenesis","volume":"11","author":"Aggeletopoulou","year":"2024","journal-title":"Front Med (Lausanne)"},{"key":"2026091509361866500_btag640-B2","doi-asserted-by":"crossref","first-page":"2794","DOI":"10.1109\/TCBB.2021.3092879","article-title":"Dynamic bayesian network learning to infer sparse models from time series gene expression data","volume":"19","author":"Ajmal","year":"2022","journal-title":"IEEE\/ACM Trans Comput Biol Bioinform"},{"key":"2026091509361866500_btag640-B3","doi-asserted-by":"crossref","first-page":"1068","DOI":"10.3390\/biomedicines9081068","article-title":"Adrenomedullin: a novel therapeutic for the treatment of inflammatory bowel disease","volume":"9","author":"Ashizuka","year":"2021","journal-title":"Biomedicines"},{"key":"2026091509361866500_btag640-B4","doi-asserted-by":"crossref","first-page":"1169","DOI":"10.1016\/j.jpsychires.2012.05.005","article-title":"Classifying responders and non-responders; does it help when there is evidence of differentially responding patient groups?","volume":"46","author":"Boessen","year":"2012","journal-title":"J Psychiatr Res"},{"key":"2026091509361866500_btag640-B5","doi-asserted-by":"crossref","first-page":"104616","DOI":"10.1016\/j.jbi.2024.104616","article-title":"Graph neural networks for clinical risk prediction based on electronic health records: a survey","volume":"151","author":"Boll","year":"2024","journal-title":"J Biomed Informatics"},{"key":"2026091509361866500_btag640-B6","doi-asserted-by":"crossref","first-page":"2442","DOI":"10.1038\/s41467-018-04696-6","article-title":"Uncovering pseudotemporal trajectories with covariates from single cell and bulk expression data","volume":"9","author":"Campbell","year":"2018","journal-title":"Nat Commun"},{"key":"2026091509361866500_btag640-B7","doi-asserted-by":"crossref","first-page":"483","DOI":"10.1038\/s41551-024-01284-6","article-title":"Simple and effective embedding model for single-cell biology built from chatgpt","volume":"9","author":"Chen","year":"2025","journal-title":"Nat Biomed Eng"},{"key":"2026091509361866500_btag640-B8","doi-asserted-by":"crossref","first-page":"1096","DOI":"10.1093\/bioinformatics\/btl056","article-title":"maSigPro: a method to identify significantly differential expression profiles in time-course microarray experiments","volume":"22","author":"Conesa","year":"2006","journal-title":"Bioinformatics"},{"key":"2026091509361866500_btag640-B9","doi-asserted-by":"crossref","first-page":"17","DOI":"10.1038\/s43586-024-00294-7","article-title":"Graph neural networks","volume":"4","author":"Corso","year":"2024","journal-title":"Nat Rev Methods Primers"},{"key":"2026091509361866500_btag640-B10","author":"Das","year":"2023"},{"key":"2026091509361866500_btag640-B11","doi-asserted-by":"crossref","first-page":"D648","DOI":"10.1093\/nar\/gkab1006","article-title":"The intact database: efficient access to fine-grained molecular interaction data","volume":"50","author":"Del Toro","year":"2022","journal-title":"Nucleic Acids Res"},{"key":"2026091509361866500_btag640-B12","first-page":"4171","author":"Devlin","year":"2019"},{"key":"2026091509361866500_btag640-B13","doi-asserted-by":"crossref","first-page":"37","DOI":"10.2202\/1544-6115.1618","article-title":"Smoothing gene expression data with network information improves consistency of regulated genes","volume":"10","author":"D\u00f8rum","year":"2011","journal-title":"Stat Appl Genet Mol Biol"},{"key":"2026091509361866500_btag640-B14","doi-asserted-by":"crossref","first-page":"btac757","DOI":"10.1093\/bioinformatics\/btac757","article-title":"Gseapy: a comprehensive package for performing gene set enrichment analysis in python","volume":"39","author":"Fang","year":"2023","journal-title":"Bioinformatics"},{"key":"2026091509361866500_btag640-B15","doi-asserted-by":"crossref","first-page":"e0305857","DOI":"10.1371\/journal.pone.0305857","article-title":"TabDEG: classifying differentially expressed genes from rna-seq data based on feature extraction and deep learning framework","volume":"19","author":"Feng","year":"2024","journal-title":"PLoS One"},{"key":"2026091509361866500_btag640-B16","doi-asserted-by":"crossref","first-page":"2336","DOI":"10.1016\/j.immuni.2022.11.002","article-title":"B cell expansion hinders the stroma-epithelium regenerative cross talk during mucosal healing","volume":"55","author":"Frede","year":"2022","journal-title":"Immunity"},{"key":"2026091509361866500_btag640-B17","doi-asserted-by":"crossref","first-page":"127136","DOI":"10.1016\/j.eswa.2025.127136","article-title":"Hybrid attentive graph neural networks for time series gene expression clustering","volume":"277","author":"Giovanoudi","year":"2025","journal-title":"Expert Syst Appl"},{"key":"2026091509361866500_btag640-B18","doi-asserted-by":"crossref","first-page":"nwaf028","DOI":"10.1093\/nsr\/nwaf028","article-title":"Foundation models in bioinformatics","volume":"12","author":"Guo","year":"2025","journal-title":"Natl Sci Rev"},{"key":"2026091509361866500_btag640-B19","doi-asserted-by":"crossref","first-page":"1318","DOI":"10.1016\/j.jaci.2021.10.036","article-title":"Ritlecitinib and brepocitinib demonstrate significant improvement in scalp alopecia areata biomarkers","volume":"149","author":"Guttman-Yassky","year":"2022","journal-title":"J Allergy Clin Immunol"},{"key":"2026091509361866500_btag640-B20","doi-asserted-by":"crossref","first-page":"1362","DOI":"10.1038\/s41598-018-19333-x","article-title":"Gene annotation bias impedes biomedical research","volume":"8","author":"Haynes","year":"2018","journal-title":"Sci Rep"},{"key":"2026091509361866500_btag640-B21","doi-asserted-by":"crossref","first-page":"244","DOI":"10.1007\/s10238-024-01508-8","article-title":"Cxcl13: a common target for immune-mediated inflammatory diseases","volume":"24","author":"Hui","year":"2024","journal-title":"Clin Exp Med"},{"key":"2026091509361866500_btag640-B22","author":"Jain","year":"2018"},{"key":"2026091509361866500_btag640-B23","doi-asserted-by":"crossref","first-page":"21342","DOI":"10.1038\/s41598-024-71864-8","article-title":"Inferring gene regulatory networks with graph convolutional network based on causal feature reconstruction","volume":"14","author":"Ji","year":"2024","journal-title":"Sci Rep"},{"key":"2026091509361866500_btag640-B24","doi-asserted-by":"crossref","first-page":"118","DOI":"10.1093\/biostatistics\/kxj037","article-title":"Adjusting batch effects in microarray expression data using empirical bayes methods","volume":"8","author":"Johnson","year":"2007","journal-title":"Biostatistics"},{"key":"2026091509361866500_btag640-B25","doi-asserted-by":"crossref","first-page":"lqad020","DOI":"10.1093\/nargab\/lqad020","article-title":"Tisa: timeseriesanalysis\u2014a pipeline for the analysis of longitudinal transcriptomics data","volume":"5","author":"Lefol","year":"2023","journal-title":"NAR Genom Bioinform"},{"key":"2026091509361866500_btag640-B26","doi-asserted-by":"crossref","first-page":"1935","DOI":"10.1016\/j.jid.2024.02.013","article-title":"Endothelial dysfunction in psoriasis: an integrative review","volume":"144","author":"Li","year":"2024","journal-title":"J Invest Dermatol"},{"key":"2026091509361866500_btag640-B27","doi-asserted-by":"crossref","first-page":"2076","DOI":"10.1109\/TCBB.2024.3442669","article-title":"Constrained pseudo-time ordering for clinical transcriptomics data","volume":"21","author":"Mathur","year":"2024","journal-title":"IEEE\/ACM Trans Comput Biol Bioinform"},{"key":"2026091509361866500_btag640-B28","doi-asserted-by":"crossref","first-page":"701","DOI":"10.1093\/intimm\/dxr077","article-title":"IL-1\u03b2 and TNF\u03b1-initiated IL-6\u2013Stat3 pathway is critical in mediating inflammatory cytokines and RANKL expression in inflammatory arthritis","volume":"23","author":"Mori","year":"2011","journal-title":"Int Immunol"},{"key":"2026091509361866500_btag640-B29","doi-asserted-by":"crossref","first-page":"352","DOI":"10.3390\/genes12030352","article-title":"Temporal dynamic methods for bulk RNA-seq time series data","volume":"12","author":"Oh","year":"2021","journal-title":"Genes (Basel)"},{"key":"2026091509361866500_btag640-B30","first-page":"494","article-title":"Carbonic anhydrase III has potential as a biomarker for experimental colitis and functions as an immune regulator by inhibiting inflammatory cytokine secretion","volume":"11","author":"Okada","year":"2022","journal-title":"Biology (Basel)"},{"key":"2026091509361866500_btag640-B31","author":"Purchase","year":"2022"},{"key":"2026091509361866500_btag640-B32","doi-asserted-by":"crossref","first-page":"R25","DOI":"10.1186\/gb-2010-11-3-r25","article-title":"A scaling normalization method for differential expression analysis of RNA-seq data","volume":"11","author":"Robinson","year":"2010","journal-title":"Genome Biol"},{"key":"2026091509361866500_btag640-B33","doi-asserted-by":"crossref","first-page":"927","DOI":"10.1038\/s41587-023-01905-6","article-title":"Predicting transcriptional outcomes of novel multigene perturbations with gears","volume":"42","author":"Roohani","year":"2024","journal-title":"Nat Biotechnol"},{"key":"2026091509361866500_btag640-B34","doi-asserted-by":"crossref","first-page":"2354","DOI":"10.1053\/j.gastro.2021.02.062","article-title":"Therapeutic interleukin-6 trans-signaling inhibition by olamkicept (sgp130Fc) in patients with active inflammatory bowel disease","volume":"160","author":"Schreiber","year":"2021","journal-title":"Gastroenterology"},{"key":"2026091509361866500_btag640-B35","doi-asserted-by":"crossref","first-page":"98","DOI":"10.1111\/cts.12538","article-title":"It\u2019s time to reverse our thinking: the reverse translation research paradigm","volume":"11","author":"Shakhnovich","year":"2018","journal-title":"Clin Transl Sci"},{"key":"2026091509361866500_btag640-B36","doi-asserted-by":"crossref","first-page":"708","DOI":"10.1038\/s41592-025-02627-0","article-title":"scNET: learning context-specific gene and cell embeddings by integrating single-cell gene expression data with protein\u2013protein interactions","volume":"22","author":"Sheinin","year":"2025","journal-title":"Nat Methods"},{"key":"2026091509361866500_btag640-B37","doi-asserted-by":"crossref","first-page":"491","DOI":"10.1038\/s43588-021-00099-8","article-title":"Modeling gene regulatory networks using neural network architectures","volume":"1","author":"Shu","year":"2021","journal-title":"Nat Comput Sci"},{"key":"2026091509361866500_btag640-B38","doi-asserted-by":"crossref","first-page":"44","DOI":"10.1016\/j.cyto.2015.10.008","article-title":"Chemokine and cytokine levels in inflammatory bowel disease patients","volume":"77","author":"Singh","year":"2016","journal-title":"Cytokine"},{"key":"2026091509361866500_btag640-B39","author":"Song","year":"2024"},{"key":"2026091509361866500_btag640-B40","doi-asserted-by":"crossref","first-page":"15545","DOI":"10.1073\/pnas.0506580102","article-title":"Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles","volume":"102","author":"Subramanian","year":"2005","journal-title":"Proc Natl Acad Sci USA"},{"key":"2026091509361866500_btag640-B41","doi-asserted-by":"crossref","first-page":"616","DOI":"10.1038\/s41586-023-06139-9","article-title":"Transfer learning enables predictions in network biology","volume":"618","author":"Theodoris","year":"2023","journal-title":"Nature"},{"key":"2026091509361866500_btag640-B42","doi-asserted-by":"crossref","first-page":"2585","DOI":"10.1038\/s41467-023-38192-3","article-title":"Chemistry-intuitive explanation of graph neural networks for molecular property prediction with substructure masking","volume":"14","author":"Wu","year":"2023","journal-title":"Nat Commun"},{"key":"2026091509361866500_btag640-B43","first-page":"75","article-title":"Secukinumab in the treatment of psoriasis: patient selection and perspectives","volume":"8","author":"Yang","year":"2018","journal-title":"Psoriasis (Auckl)"},{"key":"2026091509361866500_btag640-B44","doi-asserted-by":"crossref","first-page":"960","DOI":"10.1038\/s41467-023-36635-5","article-title":"Batch alignment of single-cell transcriptomics data using deep metric learning","volume":"14","author":"Yu","year":"2023","journal-title":"Nat Commun"},{"key":"2026091509361866500_btag640-B45","doi-asserted-by":"crossref","first-page":"120","DOI":"10.1016\/j.gpb.2019.09.006","article-title":"Htftarget: a comprehensive database for regulations of human transcription factors and their targets","volume":"18","author":"Zhang","year":"2020","journal-title":"Genomics Proteomics Bioinformatics"},{"key":"2026091509361866500_btag640-B46","doi-asserted-by":"crossref","first-page":"102327","DOI":"10.1016\/j.sbi.2021.102327","article-title":"Graph neural network approaches for drug-target interactions","volume":"73","author":"Zhang","year":"2022","journal-title":"Curr Opin Struct Biol"},{"key":"2026091509361866500_btag640-B47","doi-asserted-by":"publisher","first-page":"3848","DOI":"10.1109\/TITS.2019.2935152","article-title":"T-GCN: a temporal graph convolutional network for traffic prediction","volume":"21","author":"Zhao","year":"2020","journal-title":"IEEE Trans Intell Transport Syst"}],"container-title":["Bioinformatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/bioinformatics\/advance-article-pdf\/doi\/10.1093\/bioinformatics\/btag640\/70818938\/btag640.pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/42\/9\/btag640\/70818938\/btag640.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/42\/9\/btag640\/70818938\/btag640.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,15]],"date-time":"2026-09-15T13:36:28Z","timestamp":1789479388000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article\/doi\/10.1093\/bioinformatics\/btag640\/8771776"}},"subtitle":[],"editor":[{"given":"Janet","family":"Kelso","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2026,8,27]]},"references-count":47,"journal-issue":{"issue":"9","published-print":{"date-parts":[[2026,8,31]]}},"URL":"https:\/\/doi.org\/10.1093\/bioinformatics\/btag640","relation":{},"ISSN":["1367-4803","1367-4811"],"issn-type":[{"value":"1367-4803","type":"print"},{"value":"1367-4811","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2026,9]]},"published":{"date-parts":[[2026,8,27]]},"article-number":"btag640"}}