{"id":"W4411150076","doi":"10.1158/2159-8290.cd-24-1488","title":"Federated Deep Learning Enables Cancer Subtyping by Proteomics","year":2025,"lang":"en","type":"article","venue":"Cancer Discovery","topic":"Advanced Proteomics Techniques and Applications","field":"Chemistry","cited_by":11,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Manitoba; Research Institute in Oncology and Hematology; McGill University; CancerCare Manitoba; Jewish General Hospital","funders":"National Cancer Institute; Agència de Gestió d'Ajuts Universitaris i de Recerca; Cancer Council Victoria; Cancer Council NSW; National Health and Medical Research Council; Horizon 2020 Framework Programme; CRIS Cancer Foundation; Medical Research Council; NSW Ministry of Health; Cancer Institute NSW; University of Sydney; Australian Cancer Research Foundation; State Government of Victoria; European Commission; Fondation du cancer du sein du Québec; Children's Medical Research; National Breast Cancer Foundation; David and Elaine Potter Foundation; AstraZeneca; Tour de Cure; U.S. Department of Health and Human Services","keywords":"Computer science; Generalizability theory; Subtyping; Replicate; Biomedicine; Machine learning; Artificial intelligence; Matching (statistics); Data mining; Bioinformatics; Medicine; Biology","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.00004617844,0.00019192,0.0001986907,0.00003459595,0.0003543896,0.0001663233,0.0001890613,0.0001312705,0.0002705819],"category_scores_gemma":[0.00002364097,0.0001967584,0.00006612196,0.0002210776,0.00006620643,0.0002677586,0.0000979173,0.0003644452,0.000003347398],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003790743,"about_ca_system_score_gemma":0.0001848932,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0005452492,"about_ca_topic_score_gemma":0.0001104742,"domain_scores_codex":[0.9989662,0.000009739066,0.0002311911,0.000392679,0.00008905424,0.0003111253],"domain_scores_gemma":[0.9995351,0.00003396502,0.00012971,0.000188306,0.00006660268,0.00004625079],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.00006930237,0.0000490061,0.01022408,0.00026958,0.00008833592,0.000001199554,0.00005832505,0.00197332,0.9567683,0.002042427,0.004239842,0.02421627],"study_design_scores_gemma":[0.0003057467,0.000004505263,0.00004621369,0.0003110913,0.00003308317,5.157783e-7,0.000152056,0.002320055,0.9210067,0.001938705,0.07359397,0.0002874066],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5674745,0.01275938,0.3948716,0.001532562,0.0001642198,0.000619857,0.0002246167,0.0006270339,0.02172624],"genre_scores_gemma":[0.944145,0.006768586,0.003183097,0.0005468499,0.0001905104,0.00369944,0.0001344813,0.00006231221,0.04126969],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3916885,"threshold_uncertainty_score":0.8023576,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.008372419420359248,"score_gpt":0.2838279326398147,"score_spread":0.2754555132194554,"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."}}