{"id":"W4407173320","doi":"10.1101/2025.01.31.635690","title":"NGSTroubleFinder: A tool for detection and quantification of contamination and kinship across human NGS data","year":2025,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Domain Adaptation and Few-Shot Learning","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Discovery Centre","funders":"","keywords":"Contamination; Kinship; Geography; Biology; Sociology; Anthropology; Ecology","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":[0.008580937,0.001771706,0.001871224,0.005622643,0.002878633,0.004040543,0.003343791,0.001570113,0.04215105],"category_scores_gemma":[0.03284803,0.001975197,0.001851842,0.004225913,0.001701374,0.002765436,0.007465288,0.003442519,0.02001414],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001192196,"about_ca_system_score_gemma":0.004426934,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003876276,"about_ca_topic_score_gemma":0.006517317,"domain_scores_codex":[0.9946861,0.001039527,0.0004645141,0.001627215,0.001846337,0.0003363274],"domain_scores_gemma":[0.9865143,0.007885518,0.001348319,0.002736774,0.0009962955,0.0005187888],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0009460674,0.0001338176,0.02653118,0.004014181,0.0008354994,0.001384488,0.001787308,0.01832129,0.02506899,0.01937582,0.608176,0.2934255],"study_design_scores_gemma":[0.0004514653,0.000134971,0.02352271,0.00128887,0.0004088369,0.003169842,0.0006235761,0.182182,0.08911563,0.1248078,0.5737046,0.0005896883],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"software","genre_scores_codex":[0.01572864,0.0007623643,0.6962866,0.000871447,0.0004649705,0.0004347918,0.08663297,0.1920853,0.006732858],"genre_scores_gemma":[0.09031677,0.0004990039,0.7371665,0.001523292,0.0001858021,0.00217367,0.09681875,0.0605379,0.01077829],"genre_candidate":"software","genre_consensus":null,"teacher_disagreement_score":0.04215105,"threshold_uncertainty_score":0.1410093,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05001443210657115,"score_gpt":0.2924693571364162,"score_spread":0.2424549250298451,"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."}}