{"id":"W4408061948","doi":"10.5220/0013076600003890","title":"Towards More Robust Transcription Factor Binding Site Classifiers Using Out-of-Distribution Data","year":2025,"lang":"en","type":"article","venue":"","topic":"Cancer-related molecular mechanisms research","field":"Biochemistry, Genetics and Molecular Biology","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"Princess Margaret Cancer Centre","funders":"","keywords":"Computer science; Transcription factor; Artificial intelligence; Factor (programming language); DNA binding site; Data mining; Pattern recognition (psychology); Promoter; Biology; Genetics; Gene; Gene expression","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":[0.004811812,0.002460771,0.003433834,0.002810475,0.0007308993,0.002548042,0.003065433,0.00281222,0.00305497],"category_scores_gemma":[0.01078035,0.0008904011,0.001940407,0.001843793,0.0006834473,0.002805992,0.002389524,0.003685443,0.004618051],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006795757,"about_ca_system_score_gemma":0.001461294,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005125436,"about_ca_topic_score_gemma":0.007165152,"domain_scores_codex":[0.9967229,0.0007880091,0.0001918227,0.0009384606,0.0009770207,0.0003818761],"domain_scores_gemma":[0.9918595,0.004898431,0.0004302281,0.001376004,0.001146447,0.0002893276],"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":[0.001694546,0.001118609,0.01547207,0.000520522,0.0006993443,0.0004494397,0.0001586231,0.1359944,0.09409171,0.005013802,0.02273545,0.7220516],"study_design_scores_gemma":[0.00004138162,0.00009072898,0.001821333,0.00002855951,0.00007215046,0.0001453812,0.00006638038,0.9722797,0.01285393,0.008910251,0.003664017,0.00002609269],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.06034483,0.001246247,0.922542,0.0005416352,0.0002160313,0.0001377654,0.004009129,0.009423617,0.001538758],"genre_scores_gemma":[0.4579406,0.0009432762,0.5042563,0.001114437,0.0005620698,0.0002983954,0.0282808,0.0012112,0.005392981],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.005125436,"threshold_uncertainty_score":0.02544761,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08339644658357633,"score_gpt":0.3513607326688582,"score_spread":0.2679642860852819,"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."}}