{"id":"W4414258706","doi":"10.1101/2025.09.09.25335438","title":"Advancing Human Population Genomics with DNA Foundation Models","year":2025,"lang":"en","type":"preprint","venue":"medRxiv","topic":"Genetic Associations and Epidemiology","field":"Biochemistry, Genetics and Molecular Biology","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto; Hospital for Sick Children","funders":"","keywords":"Genomics; Human genetic variation; Population; Haplotype; Human genome; Functional genomics; Locus (genetics); Human genetics","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"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":[0.009536082,0.0007434577,0.001033941,0.002007576,0.0007337292,0.002361726,0.001925593,0.001406922,0.004785649],"category_scores_gemma":[0.02940275,0.000630211,0.001711333,0.001742162,0.001582094,0.00189979,0.002386468,0.002274547,0.000653312],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001389652,"about_ca_system_score_gemma":0.001955268,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.009937737,"about_ca_topic_score_gemma":0.007830742,"domain_scores_codex":[0.995683,0.00320821,0.00007613734,0.0005543075,0.0003561499,0.0001221947],"domain_scores_gemma":[0.9856388,0.01117224,0.0007375558,0.001457093,0.0006318283,0.0003625829],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0001009512,0.00006031853,0.01641818,0.000142591,0.0004193125,0.0003546425,0.0004153432,0.3708204,0.0009884072,0.5766668,0.00411889,0.02949421],"study_design_scores_gemma":[0.00003634083,0.00003179725,0.001421262,0.00004088505,0.00003983464,0.00008679261,0.00007144917,0.4922955,0.0001893981,0.5000932,0.005667467,0.00002614174],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.0211333,0.0002476827,0.9727447,0.001163258,0.00007600636,0.00004971018,0.0009826147,0.0004834925,0.003119131],"genre_scores_gemma":[0.4863355,0.0007328304,0.5053845,0.0008369992,0.0002105445,0.0004989821,0.002070767,0.0005232681,0.00340641],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.009937737,"threshold_uncertainty_score":0.05043226,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01781344207544698,"score_gpt":0.2864365450727414,"score_spread":0.2686231029972944,"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."}}