{"id":"W2737578470","doi":"10.1186/s12859-017-1758-x","title":"Exploring general-purpose protein features for distinguishing enzymes and non-enzymes within the twilight zone","year":2017,"lang":"en","type":"article","venue":"BMC Bioinformatics","topic":"Machine Learning in Bioinformatics","field":"Biochemistry, Genetics and Molecular Biology","cited_by":13,"is_retracted":false,"has_abstract":true,"ca_institutions":"Carleton University","funders":"European Regional Development Fund; Dirección General de Asuntos del Personal Académico, Universidad Nacional Autónoma de México; Fundação para a Ciência e a Tecnologia; Universitat de València","keywords":"Benchmark (surveying); Proteome; Computational biology; Computer science; Protein function prediction; Structural bioinformatics; Protein structure database; Sequence alignment; Similarity (geometry); Structural alignment; Protein methods; Protein sequencing; Annotation; Protein family; Protein function; Function (biology); Structural similarity; Sequence (biology); Protein structure; Multiple sequence alignment; Artificial intelligence; Biology; Bioinformatics; Peptide sequence; Sequence database; Genetics; Gene; Biochemistry","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":[0.0006803446,0.0007951349,0.000794997,0.003106206,0.0003212693,0.0009244379,0.000674842,0.0005670741,0.0009518077],"category_scores_gemma":[0.001365501,0.0001087249,0.0006361605,0.002640852,0.0004046359,0.0007750807,0.0009598343,0.0006168866,0.0006657724],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004242001,"about_ca_system_score_gemma":0.0006164032,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002392562,"about_ca_topic_score_gemma":0.003087933,"domain_scores_codex":[0.9997401,0.00002702825,0.00002057794,0.00007815628,0.0000894554,0.00004460228],"domain_scores_gemma":[0.9992954,0.0002359547,0.0001783362,0.00007653974,0.0001353014,0.00007849245],"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":[0.001937827,0.001012165,0.1920477,0.001608174,0.0003257245,0.0006792374,0.0003109205,0.08006486,0.1918419,0.005561159,0.01536062,0.5092497],"study_design_scores_gemma":[0.0001246595,0.001018365,0.1045273,0.00007411643,0.0001368894,0.0009032851,0.000643404,0.7857097,0.08286601,0.007978571,0.01591829,0.00009950019],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8879244,0.001936456,0.09891205,0.0001652844,0.00003606869,0.0001141323,0.005326194,0.002947795,0.002637687],"genre_scores_gemma":[0.9167237,0.0006852045,0.06341241,0.0000661611,0.00002016382,0.0001219154,0.01757121,0.0002002442,0.001199061],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.003106206,"threshold_uncertainty_score":0.004757345,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04100894492899753,"score_gpt":0.2774749193984699,"score_spread":0.2364659744694724,"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."}}