{"id":"W4413856438","doi":"10.1101/2025.08.26.672448","title":"A Transparent and Generalizable Deep Learning Framework for Genomic Ancestry Prediction","year":2025,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Genetic Associations and Epidemiology","field":"Biochemistry, Genetics and Molecular Biology","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Montreal Heart Institute; Université de Montréal","funders":"Alliance de recherche numérique du Canada; Institut de Valorisation des Données","keywords":"Deep learning; Artificial intelligence; Computer science; Computational biology; Data science; Machine learning; Evolutionary biology; Biology","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001866474,0.0008176449,0.0005873137,0.0006127973,0.0004089983,0.0009707073,0.001716016,0.001066609,0.001340356],"category_scores_gemma":[0.004623297,0.0004949109,0.000732436,0.0005163889,0.0009629793,0.001174804,0.002033103,0.002797907,0.0003250938],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00120805,"about_ca_system_score_gemma":0.001533972,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01280621,"about_ca_topic_score_gemma":0.0140159,"domain_scores_codex":[0.999539,0.0001533144,0.00001799227,0.0001439852,0.00009414205,0.00005162437],"domain_scores_gemma":[0.9989516,0.0005080165,0.0001148569,0.0001620851,0.0001825377,0.00008084849],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0001220762,0.00008591991,0.002741683,0.00003913013,0.00007494146,0.0001295785,0.0001077594,0.896968,0.003831865,0.01470528,0.002823985,0.0783698],"study_design_scores_gemma":[0.000003980535,0.00000736829,0.00008750441,0.000003132372,0.000002527606,0.000005490169,0.00000269975,0.9926205,0.0003002884,0.006796691,0.0001673955,0.000002480038],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.03230285,0.0001963125,0.9642665,0.0006461655,0.00003719436,0.00003298401,0.0002631279,0.001434883,0.0008198547],"genre_scores_gemma":[0.7074287,0.0002781932,0.2862046,0.0006712141,0.00009152379,0.000156092,0.00100061,0.0002492772,0.003919841],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01280621,"threshold_uncertainty_score":0.02546334,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02180367979154605,"score_gpt":0.2596691222695013,"score_spread":0.2378654424779552,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}