{"id":"W4408556965","doi":"10.1016/j.xpro.2025.103681","title":"Protocol for assessing distances in pathway space for classifier feature sets from machine learning methods","year":2025,"lang":"en","type":"article","venue":"STAR Protocols","topic":"Machine Learning in Bioinformatics","field":"Biochemistry, Genetics and Molecular Biology","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Canada's Michael Smith Genome Sciences Centre","funders":"Fundação Araucária; Conselho Nacional de Desenvolvimento Científico e Tecnológico; Coordenação de Aperfeiçoamento de Pessoal de Nível Superior; National Cancer Institute; National Institutes of Health; Fundacion Araucaria","keywords":"Computer science; Classifier (UML); Artificial intelligence; Protocol (science); Machine learning; Feature vector; Pattern recognition (psychology); 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":[0.0009713758,0.0002707083,0.0003129737,0.00007947169,0.0001901612,0.0001566179,0.0002945417,0.0002966938,0.00001494468],"category_scores_gemma":[0.0009729923,0.0002329328,0.0001358527,0.0001553959,0.00005701902,0.00001775168,0.0001310582,0.0003484343,8.58525e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004952324,"about_ca_system_score_gemma":0.0002345653,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001854524,"about_ca_topic_score_gemma":0.00009037786,"domain_scores_codex":[0.9984285,0.0002648258,0.0003847996,0.0004271245,0.0001193394,0.0003754702],"domain_scores_gemma":[0.999002,0.0001905239,0.0002686832,0.0003645225,0.0001215567,0.0000527441],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"not_applicable","study_design_scores_codex":[0.006890264,0.0007206059,0.07235207,0.008340811,0.0004475388,0.000005730092,0.001106036,0.001855048,0.5518442,0.004744208,0.0236764,0.3280171],"study_design_scores_gemma":[0.003473718,0.0003433673,0.001295148,0.0004016286,0.000009147591,6.091585e-7,0.0001121112,0.02241011,0.07365403,0.001856694,0.8961495,0.0002939788],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"protocol","genre_gemma":"protocol","genre_scores_codex":[0.0005546442,0.00001829398,0.2885019,0.0004598512,0.00005070195,0.708072,0.0001125963,0.00004179533,0.002188204],"genre_scores_gemma":[0.000387148,2.484229e-7,0.3198231,0.000131792,0.00007247601,0.677793,0.0002226797,0.00002634475,0.001543224],"genre_candidate":"protocol","genre_consensus":"protocol","teacher_disagreement_score":0.8724731,"threshold_uncertainty_score":0.9498726,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03598877127463362,"score_gpt":0.4423363084063617,"score_spread":0.4063475371317281,"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."}}