{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,1]],"date-time":"2026-10-01T05:32:09Z","timestamp":1790832729972,"version":"4.1.0"},"reference-count":35,"publisher":"Oxford University Press (OUP)","issue":"Supplement_1","license":[{"start":{"date-parts":[[2026,7,7]],"date-time":"2026-07-07T00:00:00Z","timestamp":1783382400000},"content-version":"vor","delay-in-days":6,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/100000054","name":"National Cancer Institute","doi-asserted-by":"publisher","award":["R01CA262296"],"award-info":[{"award-number":["R01CA262296"]}],"id":[{"id":"10.13039\/100000054","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,7,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Motivation<\/jats:title>\n                    <jats:p>Models deployed for genomic prediction of diseases perform unevenly across populations, limiting clinical utility. Two factors drive this limitation: large imbalances in sample availability across ancestry groups and non-stationarity of genotype\u2013phenotype effect sizes across the ancestry continuum. While tabular foundation models with in-context learning (ICL) have shown strong sample efficiency in other domains, their effectiveness for genotype-to-phenotype prediction and their robustness to ancestry-driven effect heterogeneity remain unclear.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>Using large, ancestrally diverse biobank data, we show that ICL-capable tabular foundation models reduce performance degradation in under-sampled ancestry groups compared to conventional supervised approaches. However, we find that prevailing models trained on existing synthetic tabular tasks fail when allele effect sizes vary across ancestry space. Treating genetic ancestry as a continuous variable, we introduce an instruction-tuning framework that exposes models to synthetic tasks with ancestry-dependent non-stationary effects. Instruction-tuned models achieve improved and more stable predictive performance across the genetic ancestry continuum, including for individuals distant from in-context exemplars in ancestry space.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Availability and implementation<\/jats:title>\n                    <jats:p>All code for instruction-tuning models, synthetic task generation, data wrangling, and model evaluation, is publicly available at https:\/\/github.com\/ai4pm\/Bridging-Ancestry-Gaps-in-Genomic-Risk-Prediction-with-Tabular-Foundation-Models. The final instruction-tuned model (ICL-NS-G2P-proto) is also released in this repository. Detailed documentation is provided, including environment setup instructions and guidelines for running various parts. The instruction-tuning task datasets are available at https:\/\/zenodo.org\/records\/18309187.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1093\/bioinformatics\/btag217","type":"journal-article","created":{"date-parts":[[2026,7,7]],"date-time":"2026-07-07T16:31:31Z","timestamp":1783441891000},"source":"Crossref","is-referenced-by-count":1,"title":["Bridging ancestry gaps in genomic risk prediction with tabular foundation models"],"prefix":"10.1093","volume":"42","author":[{"given":"Anirban","family":"Das","sequence":"first","affiliation":[{"name":"Department of Genetics, Genomics & Informatics, The University of Tennessee Health Science Center , Memphis, TN 38163,","place":["United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yan","family":"Cui","sequence":"additional","affiliation":[{"name":"Department of Genetics, Genomics & Informatics, The University of Tennessee Health Science Center , Memphis, TN 38163,","place":["United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"286","published-online":{"date-parts":[[2026,7,7]]},"reference":[{"key":"2026071008221701600_btag217-B1","first-page":"57125","author":"Bai"},{"key":"2026071008221701600_btag217-B2","doi-asserted-by":"crossref","first-page":"181","DOI":"10.1146\/annurev-genom-091416-035517","article-title":"The future of genomic studies must be globally representative: perspectives from page","volume":"20","author":"Bien","year":"2019","journal-title":"Annu Rev Genomics Hum Genet"},{"key":"2026071008221701600_btag217-B3","first-page":"1877","article-title":"Language models are few-shot learners","volume":"33","author":"Brown","year":"2020","journal-title":"Adv Neural Inf Process Syst"},{"key":"2026071008221701600_btag217-B4","doi-asserted-by":"crossref","first-page":"18878","DOI":"10.52202\/068431-1371","article-title":"Data distributional properties drive emergent in-context learning in transformers","volume":"35","author":"Chan","year":"2022","journal-title":"Adv Neural Inf Process Syst"},{"key":"2026071008221701600_btag217-B5","doi-asserted-by":"publisher","first-page":"340","DOI":"10.1038\/s41586-023-06957-x","article-title":"Genomic data in the all of us research program","volume":"627","author":"Choi","year":"2024","journal-title":"Nature"},{"key":"2026071008221701600_btag217-B6","doi-asserted-by":"publisher","first-page":"2759","DOI":"10.1038\/s41596-020-0353-1","article-title":"Tutorial: a guide to performing polygenic risk score analyses","volume":"15","author":"Choi","year":"2020","journal-title":"Nat Protoc"},{"key":"2026071008221701600_btag217-B7","doi-asserted-by":"publisher","first-page":"172","DOI":"10.1038\/s41746-025-01564-8","article-title":"Digital pathways connecting social and biological factors to health outcomes and equity","volume":"8","author":"Cui","year":"2025","journal-title":"NPJ Digit Med"},{"key":"2026071008221701600_btag217-B8","first-page":"1","article-title":"When no paths lead to Rome: benchmarking systematic neural relational reasoning","author":"Das","year":"2025","journal-title":"et al. (eds), Advances in Neural Information Processing Systems,"},{"key":"2026071008221701600_btag217-B9","doi-asserted-by":"publisher","first-page":"774","DOI":"10.1038\/s41586-023-06079-4","article-title":"Polygenic scoring accuracy varies across the genetic ancestry continuum","volume":"618","author":"Ding","year":"2023","journal-title":"Nature"},{"key":"2026071008221701600_btag217-B10","doi-asserted-by":"publisher","first-page":"3328","DOI":"10.1038\/s41467-019-11112-0","article-title":"Analysis of polygenic risk score usage and performance in diverse human populations","volume":"10","author":"Duncan","year":"2019","journal-title":"Nat Commun"},{"key":"2026071008221701600_btag217-B11","doi-asserted-by":"publisher","first-page":"6444","DOI":"10.1038\/s41467-020-18918-3","article-title":"Deep transfer learning for reducing health care disparities arising from biomedical data inequality","volume":"11","author":"Gao","year":"2020","journal-title":"Nat Commun"},{"key":"2026071008221701600_btag217-B12","doi-asserted-by":"publisher","first-page":"76","DOI":"10.1186\/s13073-024-01345-0","article-title":"Optimizing clinico-genomic disease prediction across ancestries: a machine learning strategy with pareto improvement","volume":"16","author":"Gao","year":"2024","journal-title":"Genome Med"},{"key":"2026071008221701600_btag217-B13","doi-asserted-by":"crossref","first-page":"153","DOI":"10.1146\/annurev-biodatasci-020722-020704","article-title":"Addressing the challenge of biomedical data inequality: an artificial intelligence perspective","volume":"6","author":"Gao","year":"2023","journal-title":"Annu Rev Biomed Data Sci"},{"key":"2026071008221701600_btag217-B14","doi-asserted-by":"publisher","first-page":"13978","DOI":"10.1038\/s41598-018-32264-x","article-title":"Analysis of racial\/ethnic representation in select basic and applied cancer research studies","volume":"8","author":"Guerrero","year":"2018","journal-title":"Sci Rep"},{"key":"2026071008221701600_btag217-B15","doi-asserted-by":"publisher","first-page":"520","DOI":"10.1038\/s41576-019-0144-0","article-title":"Genomics of disease risk in globally diverse populations","volume":"20","author":"Gurdasani","year":"2019","journal-title":"Nat Rev Genet"},{"key":"2026071008221701600_btag217-B16","doi-asserted-by":"publisher","first-page":"815","DOI":"10.1038\/s41588-024-02064-3","article-title":"Genome-wide association study meta-analysis provides insights into the etiology of heart failure and its subtypes","volume":"57","author":"Henry","year":"2025","journal-title":"Nat Genet"},{"key":"2026071008221701600_btag217-B17","doi-asserted-by":"publisher","first-page":"319","DOI":"10.1038\/s41586-024-08328-6","article-title":"Accurate predictions on small data with a tabular foundation model","volume":"637","author":"Hollmann","year":"2025","journal-title":"Nature"},{"key":"2026071008221701600_btag217-B18","doi-asserted-by":"publisher","first-page":"8","DOI":"10.1038\/s41576-023-00637-2","article-title":"Principles and methods for transferring polygenic risk scores across global populations","volume":"25","author":"Kachuri","year":"2024","journal-title":"Nat Rev Genet"},{"key":"2026071008221701600_btag217-B19","doi-asserted-by":"publisher","first-page":"246","DOI":"10.1038\/s41576-025-00900-8","article-title":"Clinical use of polygenic risk scores: current status, barriers and future directions","volume":"27","author":"Kullo","year":"2026","journal-title":"Nat Rev Genet"},{"key":"2026071008221701600_btag217-B20","doi-asserted-by":"publisher","first-page":"829","DOI":"10.1038\/s41588-025-02140-2","article-title":"Common-variant and rare-variant genetic architecture of heart failure across the allele-frequency spectrum","volume":"57","author":"Lee","year":"2025","journal-title":"Nat Genet"},{"key":"2026071008221701600_btag217-B21","doi-asserted-by":"publisher","first-page":"973","DOI":"10.1016\/j.ajhg.2022.05.002","article-title":"Lessons learned from the emerge network: balancing genomics in discovery and practice","volume":"109","author":"Lennon","year":"2022","journal-title":"Am J Hum Genet"},{"key":"2026071008221701600_btag217-B22","doi-asserted-by":"publisher","first-page":"995","DOI":"10.1016\/j.tig.2021.06.004","article-title":"Genetic prediction of complex traits with polygenic scores: a statistical review","volume":"37","author":"Ma","year":"2021","journal-title":"Trends Genet"},{"key":"2026071008221701600_btag217-B23","first-page":"49","article-title":"On the generalised distance in statistics","volume":"2","author":"Mahalanobis","year":"1936","journal-title":"Proc Natl Inst Sci India"},{"key":"2026071008221701600_btag217-B24","doi-asserted-by":"publisher","first-page":"242","DOI":"10.1038\/s41588-020-0580-y","article-title":"The GWAS diversity monitor tracks diversity by disease in real time","volume":"52","author":"Mills","year":"2020","journal-title":"Nat Genet"},{"key":"2026071008221701600_btag217-B25","doi-asserted-by":"publisher","first-page":"93","DOI":"10.1186\/s40246-024-00664-y","article-title":"Polygenic risk score portability for common diseases across genetically diverse populations","volume":"18","author":"Moreno-Grau","year":"2024","journal-title":"Hum Genomics"},{"key":"2026071008221701600_btag217-B26","doi-asserted-by":"publisher","first-page":"589","DOI":"10.1016\/j.cell.2019.08.051","article-title":"Genome-wide association studies in ancestrally diverse populations","volume":"179","author":"Peterson","year":"2019","journal-title":"Cell"},{"key":"2026071008221701600_btag217-B27","doi-asserted-by":"publisher","first-page":"12","DOI":"10.1016\/j.ajhg.2021.11.008","article-title":"Portability of 245 polygenic scores when derived from the UK Biobank and applied to 9 ancestry groups from the same cohort","volume":"109","author":"Priv\u00e9","year":"2022","journal-title":"Am J Hum Genet"},{"key":"2026071008221701600_btag217-B28","first-page":"12","article-title":"Omics-based cancer prognosis across ethnic groups: from feature engineering to disparity detection and mitigation","volume":"6","author":"Sharma","year":"2026","journal-title":"IEEE Trans Artif Intell"},{"key":"2026071008221701600_btag217-B29","doi-asserted-by":"publisher","first-page":"410","DOI":"10.1038\/s41588-023-01314-0","article-title":"Multi-ancestry genome-wide association analyses improve resolution of genes and pathways influencing lung function and chronic obstructive pulmonary disease risk","volume":"55","author":"Shrine","year":"2023","journal-title":"Nat Genet"},{"key":"2026071008221701600_btag217-B30","doi-asserted-by":"publisher","first-page":"26","DOI":"10.1016\/j.cell.2019.02.048","article-title":"The missing diversity in human genetic studies","volume":"177","author":"Sirugo","year":"2019","journal-title":"Cell"},{"key":"2026071008221701600_btag217-B31","doi-asserted-by":"publisher","first-page":"1679","DOI":"10.1038\/s41591-022-01891-3","article-title":"Large-scale genome-wide association study of coronary artery disease in genetically diverse populations","volume":"28","author":"Tcheandjieu","year":"2022","journal-title":"Nat Med"},{"key":"2026071008221701600_btag217-B32","doi-asserted-by":"publisher","first-page":"3865","DOI":"10.1038\/s41467-020-17719-y","article-title":"Theoretical and empirical quantification of the accuracy of polygenic scores in ancestry divergent populations","volume":"11","author":"Wang","year":"2020","journal-title":"Nat Commun"},{"key":"2026071008221701600_btag217-B33","doi-asserted-by":"publisher","first-page":"100408","DOI":"10.1016\/j.xgen.2023.100408","article-title":"Polygenic prediction across populations is influenced by ancestry, genetic architecture, and methodology","volume":"3","author":"Wang","year":"2023","journal-title":"Cell Genom"},{"key":"2026071008221701600_btag217-B34","doi-asserted-by":"publisher","author":"Wei","DOI":"10.48550\/arXiv.2109.01652,"},{"key":"2026071008221701600_btag217-B35","doi-asserted-by":"publisher","first-page":"100192","DOI":"10.1016\/j.xgen.2022.100192","article-title":"Global biobank meta-analysis initiative: powering genetic discovery across human disease","volume":"2","author":"Zhou","year":"2022","journal-title":"Cell Genom"}],"container-title":["Bioinformatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/42\/Supplement_1\/btag217\/68709312\/btag217.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/42\/Supplement_1\/btag217\/68709312\/btag217.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,10]],"date-time":"2026-07-10T12:22:25Z","timestamp":1783686145000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article\/doi\/10.1093\/bioinformatics\/btag217\/8726311"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,1]]},"references-count":35,"journal-issue":{"issue":"Supplement_1","published-print":{"date-parts":[[2026,7,1]]}},"URL":"https:\/\/doi.org\/10.1093\/bioinformatics\/btag217","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":[[2026,7,1]]},"article-number":"btag217"}}