{"id":"W4283331250","doi":"10.1101/2022.06.20.22276419","title":"Evaluating multiple next-generation sequencing derived tumor features to accurately predict DNA mismatch repair status","year":2022,"lang":"en","type":"preprint","venue":"medRxiv","topic":"Genetic factors in colorectal cancer","field":"Medicine","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"Lunenfeld-Tanenbaum Research Institute; Mount Sinai Hospital; University of Toronto; Ontario Institute for Cancer Research; Alberta Health Services; University of Calgary","funders":"National Health and Medical Research Council; Canadian Institutes of Health Research; Medical Research Council; State Government of Victoria; University of Melbourne","keywords":"Microsatellite instability; Indel; DNA mismatch repair; Lasso (programming language); Feature (linguistics); Computational biology; MLH1; Artificial intelligence; Computer science; Cancer; Genetics; Biology; Gene; Colorectal cancer; Microsatellite; Single-nucleotide polymorphism","routes":{"ca_aff":true,"ca_fund":true,"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":[0.004556523,0.000945053,0.0008548207,0.001466111,0.0002831948,0.0009990181,0.0004180492,0.0006055035,0.000945975],"category_scores_gemma":[0.004893656,0.0002674951,0.001079353,0.0004693108,0.0002165562,0.000548208,0.000627782,0.0007567097,0.0004359679],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003597025,"about_ca_system_score_gemma":0.0003603057,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001516539,"about_ca_topic_score_gemma":0.002502055,"domain_scores_codex":[0.9988049,0.0005278014,0.00008016717,0.000293253,0.0002110477,0.00008275275],"domain_scores_gemma":[0.9974169,0.001630558,0.0003659375,0.000154159,0.0002983538,0.0001340819],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.001996237,0.0005954564,0.691256,0.0002339096,0.002136723,0.0002714873,0.000116698,0.1119018,0.0404189,0.0002832061,0.002032579,0.148757],"study_design_scores_gemma":[0.00005979974,0.0008276051,0.09985159,0.0000480387,0.0005381427,0.0004023321,0.00006498149,0.8836026,0.01280204,0.0006644846,0.00108164,0.00005678802],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9480607,0.001283589,0.0471471,0.0003681894,0.00005414254,0.00008526286,0.001667546,0.000456702,0.0008766726],"genre_scores_gemma":[0.9807249,0.0001305497,0.01697724,0.0001162636,0.00003168027,0.00004068553,0.00157969,0.00002755056,0.0003714409],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.004556523,"threshold_uncertainty_score":0.0240975,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2230233344905034,"score_gpt":0.381160473290402,"score_spread":0.1581371387998986,"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."}}