{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.01112447,0.00282651,0.001531632,0.00399492,0.002813153,0.002940101,0.003953743,0.002014169,0.1369198],"category_scores_gemma":[0.05966099,0.002591652,0.002989356,0.003208172,0.001735852,0.002160462,0.004786339,0.008461863,0.05963205],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002028575,"about_ca_system_score_gemma":0.006183521,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00315058,"about_ca_topic_score_gemma":0.004691309,"domain_scores_codex":[0.9897002,0.002889262,0.001797157,0.001618925,0.003521426,0.0004729846],"domain_scores_gemma":[0.9577809,0.01728928,0.001477086,0.00937566,0.01296519,0.001111867],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.002541284,0.0007804575,0.005524859,0.004288708,0.0006178365,0.0008393654,0.001544035,0.008006711,0.06556492,0.04524314,0.5206108,0.3444379],"study_design_scores_gemma":[0.0009800623,0.0008150302,0.008825736,0.0008850555,0.0002329262,0.000891304,0.0005000359,0.02281948,0.09887413,0.08318044,0.7813602,0.0006355809],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"methods","genre_gemma":"protocol","genre_scores_codex":[0.003323721,0.0002895533,0.895106,0.0007295862,0.000669017,0.0082441,0.04097925,0.04367481,0.006983874],"genre_scores_gemma":[0.008387503,0.0002032415,0.8999217,0.0005779269,0.0001027808,0.03778979,0.03138151,0.01427711,0.007358484],"genre_candidate":"protocol","genre_consensus":null,"teacher_disagreement_score":0.1369198,"threshold_uncertainty_score":0.4580423,"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."}}