{"id":"W2076460489","doi":"10.1186/1471-2105-10-414","title":"Modular prediction of protein structural classes from sequences of twilight-zone identity with predicting sequences","year":2009,"lang":"en","type":"article","venue":"BMC Bioinformatics","topic":"Machine Learning in Bioinformatics","field":"Biochemistry, Genetics and Molecular Biology","cited_by":96,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Computer science; Benchmark (surveying); Artificial intelligence; Support vector machine; Modular design; Similarity (geometry); Feature selection; Feature (linguistics); Pattern recognition (psychology); Class (philosophy); Sequence (biology); Contrast (vision); Data mining; Machine learning; Biology; Image (mathematics)","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":[0.0006842242,0.0007177558,0.0004384232,0.001337361,0.0002710791,0.0005562503,0.0006225752,0.0004578147,0.001415195],"category_scores_gemma":[0.00158909,0.0001217889,0.0006609767,0.0007612478,0.0002071482,0.0005808901,0.0005053002,0.0007411414,0.0009402703],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003408004,"about_ca_system_score_gemma":0.0007477426,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001794824,"about_ca_topic_score_gemma":0.002832456,"domain_scores_codex":[0.9996303,0.00004640136,0.00002503166,0.0001223527,0.0001269734,0.00004883596],"domain_scores_gemma":[0.9989783,0.0003008709,0.0002262055,0.0001134323,0.0002976964,0.00008356666],"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":[0.001225787,0.0006638173,0.1554181,0.0003696885,0.000221235,0.0002547988,0.0001340525,0.08986264,0.0982464,0.001359177,0.01152406,0.6407202],"study_design_scores_gemma":[0.00004179334,0.0003963003,0.03597361,0.00004236151,0.00006981428,0.0002999974,0.00008138883,0.9179043,0.03967698,0.00211754,0.003369877,0.00002615425],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7948865,0.0008885558,0.1947156,0.0002659296,0.00006936587,0.000179908,0.002471041,0.003818751,0.002704369],"genre_scores_gemma":[0.8877664,0.0002118871,0.1034728,0.00009911836,0.00005609793,0.0001305548,0.006867981,0.00008492981,0.001310283],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.001794824,"threshold_uncertainty_score":0.004734278,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01020080503328411,"score_gpt":0.2450921923098636,"score_spread":0.2348913872765795,"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."}}