{"id":"W1601201751","doi":"10.1007/11510888_33","title":"Prediction of Secondary Protein Structure Content from Primary Sequence Alone – A Feature Selection Based Approach","year":2005,"lang":"en","type":"book-chapter","venue":"Lecture notes in computer science","topic":"Machine Learning in Bioinformatics","field":"Biochemistry, Genetics and Molecular Biology","cited_by":16,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Alberta","funders":"","keywords":"Feature selection; Computer science; Representation (politics); Artificial intelligence; Set (abstract data type); Pattern recognition (psychology); Feature (linguistics); Protein function prediction; Selection (genetic algorithm); Sequence (biology); Protein secondary structure; Function (biology); Data mining; Machine learning; Protein function; 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":[0.0005506599,0.0007719256,0.001267815,0.001733637,0.0003032633,0.0004448364,0.0005195569,0.0005354066,0.001122953],"category_scores_gemma":[0.0005480946,0.000260935,0.001038322,0.001142479,0.0001908035,0.0004301907,0.0003158383,0.0004319913,0.0009069844],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001956387,"about_ca_system_score_gemma":0.0002746225,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0009768021,"about_ca_topic_score_gemma":0.001187767,"domain_scores_codex":[0.9998217,0.00002717479,0.00001638821,0.00005010904,0.00005848036,0.00002621379],"domain_scores_gemma":[0.999453,0.0002835113,0.00005148394,0.00003934692,0.0001401891,0.00003244513],"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":[0.0009256864,0.0004323623,0.01061296,0.000193204,0.0002401949,0.000454923,0.00004622847,0.03059454,0.3238327,0.0004000723,0.002692258,0.6295749],"study_design_scores_gemma":[0.0000927009,0.0007670447,0.03209224,0.00002309014,0.0003650792,0.0007521477,0.00005513423,0.8923807,0.06957815,0.002078164,0.001755026,0.00006048453],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.43725,0.001347097,0.5550163,0.0001116961,0.00006105266,0.0001480752,0.001122542,0.003139822,0.001803422],"genre_scores_gemma":[0.7631457,0.0006180109,0.2293662,0.00007783539,0.00009511077,0.0001341839,0.003769819,0.0001953675,0.002597814],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.001733637,"threshold_uncertainty_score":0.003756642,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01556352488451747,"score_gpt":0.2160696792375443,"score_spread":0.2005061543530268,"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."}}