{"id":"W2104873783","doi":"10.1093/bioinformatics/btm475","title":"PFRES: protein fold classification by using evolutionary information and predicted secondary structure","year":2007,"lang":"en","type":"article","venue":"Bioinformatics","topic":"Machine Learning in Bioinformatics","field":"Biochemistry, Genetics and Molecular Biology","cited_by":122,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Protein tertiary structure; Protein secondary structure; Computer science; Pseudo amino acid composition; Similarity (geometry); Classifier (UML); Pattern recognition (psychology); Protein Data Bank (RCSB PDB); Fold (higher-order function); Protein superfamily; Protein structure prediction; Representation (politics); Feature vector; Structural Classification of Proteins database; Protein structure; Sequence (biology); Structural alignment; Data mining; Algorithm; Artificial intelligence; Sequence alignment; Biology; Peptide sequence; Amino acid; Genetics","routes":{"ca_aff":true,"ca_fund":true,"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":[0.001762268,0.002271978,0.0009641612,0.004458984,0.0005113001,0.0007400129,0.001837098,0.0008691765,0.009428702],"category_scores_gemma":[0.004529552,0.0003460553,0.000867346,0.001803791,0.0003854776,0.001712441,0.0008233617,0.0006871868,0.008114785],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004282333,"about_ca_system_score_gemma":0.0006279136,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001019288,"about_ca_topic_score_gemma":0.001064692,"domain_scores_codex":[0.9989383,0.0001826138,0.00008272678,0.0002290374,0.0004907784,0.00007662265],"domain_scores_gemma":[0.9988106,0.0004344647,0.0002461939,0.0001730959,0.0002654665,0.00007002835],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.001316377,0.0002658779,0.017256,0.001452574,0.0002474306,0.0005319269,0.0001342109,0.02348853,0.03655168,0.003562001,0.1166196,0.7985737],"study_design_scores_gemma":[0.0001741587,0.00058609,0.01742295,0.0002586412,0.0001008752,0.002520771,0.000116313,0.8369206,0.0848482,0.009892185,0.0469894,0.0001698311],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.08022684,0.002194039,0.6666846,0.0005556273,0.0002571969,0.0005764412,0.03260345,0.2111149,0.005786968],"genre_scores_gemma":[0.2992299,0.001020059,0.6056753,0.0001601769,0.0002349738,0.0006459322,0.08367693,0.003446512,0.005910262],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.009428702,"threshold_uncertainty_score":0.03154218,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.006359821080226958,"score_gpt":0.233839707426869,"score_spread":0.2274798863466421,"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."}}