{"id":"W4312213693","doi":"10.1101/2022.12.22.521582","title":"LegNet: a best-in-class deep learning model for short DNA regulatory regions","year":2022,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"RNA and protein synthesis mechanisms","field":"Biochemistry, Genetics and Molecular Biology","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"Russian Science Foundation","keywords":"Computer science; Python (programming language); Artificial intelligence; Deep learning; Computational biology; Data mining; Programming language; Biology","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":[0.0005897861,0.0009176513,0.0004692331,0.0005200154,0.0002419937,0.0007316487,0.00133854,0.001173005,0.004067821],"category_scores_gemma":[0.001129879,0.0003731259,0.0005365169,0.0005026255,0.0005457618,0.0008750807,0.0006755341,0.001223045,0.001649317],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009728233,"about_ca_system_score_gemma":0.0008758954,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006418015,"about_ca_topic_score_gemma":0.009847592,"domain_scores_codex":[0.999832,0.00002974777,0.000006155483,0.00006950286,0.00004056092,0.00002200786],"domain_scores_gemma":[0.9998009,0.00008991262,0.00001737479,0.00002720129,0.00004347621,0.00002108936],"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":[0.0002668539,0.0001166536,0.001113585,0.0001494199,0.00006234274,0.0001242112,0.00004142681,0.8457375,0.01510349,0.0170358,0.01770692,0.1025417],"study_design_scores_gemma":[0.000009224298,0.00001567089,0.00007309386,0.000007413628,0.000004255125,0.00001124661,0.000002341841,0.9888263,0.003041854,0.006000279,0.002003609,0.000004721787],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.06902494,0.001578922,0.8988257,0.001086282,0.0002382851,0.0001061275,0.005261416,0.01471048,0.009167911],"genre_scores_gemma":[0.604605,0.001184019,0.3410508,0.001069384,0.000158876,0.000445941,0.01918952,0.002359552,0.02993695],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.006418015,"threshold_uncertainty_score":0.01360816,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01832860580655825,"score_gpt":0.2330997595190344,"score_spread":0.2147711537124762,"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."}}