{"id":"W4406018731","doi":"10.1016/j.ccell.2024.12.002","title":"Classification of non-TCGA cancer samples to TCGA molecular subtypes using compact feature sets","year":2025,"lang":"en","type":"article","venue":"Cancer Cell","topic":"Radiomics and Machine Learning in Medical Imaging","field":"Medicine","cited_by":28,"is_retracted":false,"has_abstract":true,"ca_institutions":"Canada's Michael Smith Genome Sciences Centre","funders":"National Human Genome Research Institute; Eisai; National Cancer Institute; National Institutes of Health; Boston Scientific Corporation; Gilead Sciences; Pfizer","keywords":"Biology; Cancer research; Feature (linguistics); Computational biology; Oncology; Internal medicine; 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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0001189891,0.0001315551,0.0003034872,0.00009394084,0.00005279214,0.00001515162,0.0001140478,0.00007983665,0.0001302265],"category_scores_gemma":[0.00005843985,0.0001156622,0.0000766064,0.0002846768,0.00005962206,0.00003431569,0.0000263218,0.0002470438,0.000002346315],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002622024,"about_ca_system_score_gemma":0.0003482748,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002456269,"about_ca_topic_score_gemma":0.00005044404,"domain_scores_codex":[0.9991488,0.00002761354,0.0001817959,0.0002508177,0.0001829804,0.0002079322],"domain_scores_gemma":[0.9993628,0.00003987078,0.00009068391,0.0002531489,0.0001217694,0.0001317419],"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":[0.0001008689,0.00006529776,0.1040774,0.0004966869,0.00009304492,0.00001014371,0.0003604463,0.003728278,0.8567796,0.0001093192,0.01757564,0.01660329],"study_design_scores_gemma":[0.002595922,0.0001279708,0.6163376,0.002371445,0.0006888289,0.000005571654,0.0004308595,0.09741408,0.1885278,0.0001302729,0.09092779,0.0004418939],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9751866,0.004145666,0.007588945,0.007307433,0.0004767834,0.0003169951,0.0000237828,0.00002862263,0.004925184],"genre_scores_gemma":[0.9930282,0.0003235481,0.003157534,0.002244415,0.0001074795,0.00001363875,0.00002002395,0.00002216459,0.001082953],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.6682518,"threshold_uncertainty_score":0.4716567,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02152446678870231,"score_gpt":0.3587207373482867,"score_spread":0.3371962705595844,"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."}}