{"id":"W3043189602","doi":"10.1038/s41598-020-68858-7","title":"Inferring disease subtypes from clusters in explanation space","year":2020,"lang":"en","type":"article","venue":"Scientific Reports","topic":"Gene expression and cancer classification","field":"Biochemistry, Genetics and Molecular Biology","cited_by":29,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University; Mila - Quebec Artificial Intelligence Institute; Montreal Neurological Institute and Hospital","funders":"Medical Research Council","keywords":"Classifier (UML); Cluster analysis; Computer science; Artificial intelligence; Biobank; Identification (biology); Machine learning; Disease; Computational biology; Data mining; Pattern recognition (psychology); Bioinformatics; Biology; Medicine","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":[0.002622647,0.001061813,0.0008283573,0.003903508,0.0007276654,0.002197544,0.001316224,0.001436862,0.002382615],"category_scores_gemma":[0.01271197,0.0003951154,0.002026451,0.002163694,0.001225465,0.002259959,0.001913779,0.002007269,0.000480635],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001613094,"about_ca_system_score_gemma":0.001880013,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01257387,"about_ca_topic_score_gemma":0.01337156,"domain_scores_codex":[0.9982865,0.0006149234,0.0001132995,0.0005221521,0.0002863431,0.0001767052],"domain_scores_gemma":[0.9908319,0.006522436,0.0007414461,0.001097081,0.0005929866,0.0002142449],"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":[0.001141664,0.0003827145,0.2075237,0.0007125135,0.0008982456,0.0008155207,0.002979369,0.2976802,0.005421848,0.1254582,0.01414417,0.3428419],"study_design_scores_gemma":[0.00004926904,0.00006716833,0.01331452,0.00007964913,0.000113981,0.0002183172,0.000501666,0.7234122,0.001462443,0.2573501,0.003380718,0.00004987586],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2057317,0.001117785,0.7818123,0.002408537,0.00004699441,0.0002046728,0.005188412,0.001685198,0.00180438],"genre_scores_gemma":[0.8236851,0.0004594158,0.1664268,0.000305985,0.00006985558,0.0001969699,0.007522619,0.0001146189,0.001218767],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01257387,"threshold_uncertainty_score":0.02500135,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01734074709196928,"score_gpt":0.2482250143554381,"score_spread":0.2308842672634689,"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."}}