{"id":"W4385237450","doi":"10.1093/bioinformatics/btad457","title":"LegNet: a best-in-class deep learning model for short DNA regulatory regions","year":2023,"lang":"en","type":"article","venue":"Bioinformatics","topic":"Genomics and Chromatin Dynamics","field":"Biochemistry, Genetics and Molecular Biology","cited_by":46,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"Ministry of Science and Higher Education of the Russian Federation; Russian Science Foundation","keywords":"Computer science; Gene regulatory network; Python (programming language); Computational biology; Scripting language; Deep learning; Artificial intelligence; Gene; Data mining; Gene expression; Biology; Genetics; Programming language","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":[0.0002283952,0.0001532234,0.000149864,0.00009805601,0.0001139241,0.00003282644,0.0001979709,0.0001837927,0.000001560411],"category_scores_gemma":[0.00006350251,0.000160265,0.0001062524,0.0001385107,0.00004486332,0.000007714258,0.0001223623,0.00009161425,0.00002587681],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000309812,"about_ca_system_score_gemma":0.00008095184,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000001853674,"about_ca_topic_score_gemma":0.00006928837,"domain_scores_codex":[0.999025,0.000009798191,0.0003506987,0.0001590311,0.0001091647,0.0003463468],"domain_scores_gemma":[0.9994596,0.00001487879,0.00007684832,0.0003181997,0.00005641121,0.00007410177],"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":[0.0001637434,0.0002307543,0.004313666,0.000907782,0.00023997,0.000009475883,0.00452659,0.8162229,0.0990476,0.008758665,0.01845108,0.04712777],"study_design_scores_gemma":[0.0003250148,0.0001100064,0.0003272249,0.00001779329,0.00001349267,0.000006113942,0.0006379855,0.9892098,0.00096094,0.0003838557,0.007792438,0.0002153831],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9345113,0.0001105413,0.0618359,0.000123027,0.0001338048,0.000417774,0.00003840698,0.00005508351,0.002774181],"genre_scores_gemma":[0.9797698,0.0003617,0.01552422,0.0001583065,0.00009435602,0.00008089282,0.0006597214,0.00004588231,0.003305103],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.1729869,"threshold_uncertainty_score":0.6535417,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0159666890867824,"score_gpt":0.2420830472253247,"score_spread":0.2261163581385423,"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."}}