{"id":"W4414023774","doi":"10.1101/2025.09.02.673402","title":"Tractor Workflow Pipeline: A Scalable Nextflow Framework for Local Ancestry-Aware Genome-Wide Association Studies","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":"Public Health Ontario; University of Toronto; Centre for Addiction and Mental Health","funders":"National Institutes of Health","keywords":"Workflow; Tractor; Pipeline (software); Scalability; Computer science; Association (psychology); Genome; Computational biology; Biology; Engineering; Database; Genetics; Gene; Operating system; Psychology","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":[0.007464631,0.001919653,0.001470682,0.002111956,0.0013146,0.003421125,0.003403309,0.001396958,0.02072264],"category_scores_gemma":[0.01260508,0.001531985,0.003196125,0.001581493,0.0009785281,0.002435271,0.003547811,0.002896395,0.007169204],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001469883,"about_ca_system_score_gemma":0.004949035,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01540749,"about_ca_topic_score_gemma":0.01410471,"domain_scores_codex":[0.9984226,0.0003889189,0.0001917306,0.000482269,0.0003402475,0.0001741175],"domain_scores_gemma":[0.996449,0.001786965,0.0002329972,0.000684371,0.0005430056,0.0003037521],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.004664865,0.000486209,0.01669969,0.002122005,0.001378701,0.001879379,0.001744013,0.09425101,0.03754665,0.06505625,0.368348,0.4058231],"study_design_scores_gemma":[0.0009468222,0.0001641427,0.004517439,0.0002927613,0.0002010929,0.0004672188,0.0001768667,0.6576748,0.029823,0.1360376,0.1692845,0.0004136822],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.003075359,0.0003406754,0.8190727,0.0005019184,0.000155914,0.0003571547,0.009559117,0.165559,0.001378095],"genre_scores_gemma":[0.05581026,0.000739348,0.8723917,0.0009081169,0.0001299789,0.001908016,0.04190234,0.02169833,0.004511848],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.02072264,"threshold_uncertainty_score":0.0693242,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0247949018602809,"score_gpt":0.2824059455972613,"score_spread":0.2576110437369805,"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."}}