{"id":"W3129155125","doi":"10.1016/j.compbiomed.2021.104258","title":"FAD-BERT: Improved prediction of FAD binding sites using pre-training of deep bidirectional transformers","year":2021,"lang":"en","type":"article","venue":"Computers in Biology and Medicine","topic":"Machine Learning in Bioinformatics","field":"Biochemistry, Genetics and Molecular Biology","cited_by":58,"is_retracted":false,"has_abstract":false,"ca_institutions":"Artificial Intelligence in Medicine (Canada)","funders":"Ministry of Science and Technology, Taiwan","keywords":"Transformer; Computer science; Artificial intelligence; Training set; Training (meteorology); Machine learning; Electrical engineering; Engineering; Voltage; Physics","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":[0.0002574016,0.00009978449,0.0002305232,0.0001024292,0.00003797614,0.000001482428,0.00005922937,0.000157581,0.00001043054],"category_scores_gemma":[0.0001509092,0.00008674248,0.00003400016,0.000126186,0.0002546932,0.000004601849,0.00004507442,0.0001121466,5.921666e-8],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00000957785,"about_ca_system_score_gemma":0.00004554922,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002190904,"about_ca_topic_score_gemma":0.00002042885,"domain_scores_codex":[0.9992517,0.00005952165,0.0003378618,0.0001663348,0.0000480815,0.0001365437],"domain_scores_gemma":[0.9996218,0.00005736648,0.0001288615,0.00009116896,0.0000658298,0.00003497308],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00005752609,0.00002153908,0.05692557,0.0001078621,0.00006718304,7.78749e-7,0.001090592,0.0002273192,0.9274507,0.000112948,0.00001839421,0.01391961],"study_design_scores_gemma":[0.0096393,0.004343121,0.1793909,0.001164226,0.0002518559,0.0006197179,0.004379337,0.4077304,0.3867422,0.0007325727,0.004266469,0.0007399365],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9370738,0.0008188283,0.06125173,0.00008363038,0.0003441281,0.00007970378,0.00001248959,0.00000616591,0.0003295845],"genre_scores_gemma":[0.9869662,0.0003066281,0.01223975,0.00008562385,0.0001225668,0.000001916936,0.0002515778,0.000006124422,0.00001963631],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.5407085,"threshold_uncertainty_score":0.3537256,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01617659853623141,"score_gpt":0.3024081556868805,"score_spread":0.2862315571506491,"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."}}