{"id":"W4367301984","doi":"10.1101/2023.04.26.538471","title":"Evaluation and optimization of sequence-based gene regulatory deep learning models","year":2023,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Genomics and Chromatin Dynamics","field":"Biochemistry, Genetics and Molecular Biology","cited_by":9,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"Institute for Information and Communications Technology Promotion; National Institutes of Health; Alliance de recherche numérique du Canada; Natural Sciences and Engineering Research Council of Canada; Ministry of Science and ICT, South Korea; Seoul National University; University of British Columbia; National Research Foundation of Korea; National Research Foundation; Michael Smith Health Research BC; Russian Science Foundation; Stem Cell Network","keywords":"Sequence (biology); Computational biology; Gene; Deep learning; Artificial intelligence; Computer science; Genetics; 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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.003927962,0.002555298,0.001377698,0.0008910631,0.0004929291,0.001320242,0.0029458,0.002390101,0.001760898],"category_scores_gemma":[0.007413529,0.0009872034,0.001357815,0.0008635399,0.001138421,0.001620773,0.001476146,0.00240616,0.0007062872],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.003224807,"about_ca_system_score_gemma":0.002608642,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01764031,"about_ca_topic_score_gemma":0.01768109,"domain_scores_codex":[0.9990065,0.0003390564,0.00006725264,0.0002880055,0.000161091,0.0001380938],"domain_scores_gemma":[0.9976427,0.001491696,0.00008225587,0.0002535934,0.0004024589,0.0001272256],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0001875601,0.00009986733,0.001531578,0.0001640725,0.00009878661,0.00003978525,0.00002315584,0.9756797,0.001095743,0.00124816,0.00217246,0.01765916],"study_design_scores_gemma":[0.00002533829,0.0000587638,0.0001694365,0.00001511044,0.00001331487,0.000007610016,0.00001227014,0.9970973,0.001327381,0.0009406436,0.0003268523,0.000006011206],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7842428,0.008196196,0.1665183,0.003252749,0.0005062966,0.000346438,0.006517999,0.01540832,0.01501082],"genre_scores_gemma":[0.8915666,0.001087719,0.09072293,0.0008520024,0.00005568526,0.0003181691,0.01097724,0.0007289794,0.003690652],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01764031,"threshold_uncertainty_score":0.03507531,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02641895900319936,"score_gpt":0.2330920031233339,"score_spread":0.2066730441201345,"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."}}