{"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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.001468388,0.0002579781,0.0003352375,0.0002087055,0.000311564,0.000482313,0.0007583287,0.0002931679,0.00000101715],"category_scores_gemma":[0.0003909795,0.000303292,0.0000418024,0.0003112445,0.0001050376,0.0005395406,0.0009141123,0.0003128744,9.726075e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007301678,"about_ca_system_score_gemma":0.0001696852,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004854492,"about_ca_topic_score_gemma":0.00001349111,"domain_scores_codex":[0.9978885,0.0001168918,0.0004856857,0.001006938,0.0002401189,0.0002618473],"domain_scores_gemma":[0.9974409,0.0001717795,0.0005451788,0.001309056,0.0004547802,0.00007832906],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"observational","study_design_scores_codex":[0.00005464682,0.0001260052,0.006927181,0.00167373,0.0001390657,0.000002559836,0.0002377252,0.00009234406,0.9429896,0.0466653,0.00005809973,0.001033718],"study_design_scores_gemma":[0.002267591,0.0001672326,0.5896915,0.0006718836,0.0001334056,5.850389e-8,0.00003800138,0.164626,0.2349055,0.00009884467,0.00636132,0.001038691],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3696283,0.0002316603,0.6287625,0.0001336372,0.0003282186,0.0006168452,0.0001472267,0.0001484484,0.000003131123],"genre_scores_gemma":[0.975921,0.00006585565,0.02375169,0.00006099599,0.00006033402,0.0001075758,0.000002266084,0.00001985886,0.00001039681],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7080841,"threshold_uncertainty_score":0.9999419,"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."}}