{"id":"W2328398973","doi":"10.2174/092986612798472910","title":"CRYSpred: Accurate Sequence-Based Protein Crystallization Propensity Prediction Using Sequence-Derived Structural Characteristics","year":2012,"lang":"en","type":"article","venue":"Protein and Peptide Letters","topic":"Machine Learning in Bioinformatics","field":"Biochemistry, Genetics and Molecular Biology","cited_by":22,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"Natural Sciences and Engineering Research Council of Canada; Killam Trusts","keywords":"Structural genomics; Computer science; Protein structure prediction; Benchmark (surveying); Protein crystallization; Context (archaeology); Test set; Genomics; Protein sequencing; Classifier (UML); Selection (genetic algorithm); In silico; Sequence (biology); Rule of thumb; Artificial intelligence; Data mining; Machine learning; Crystallization; Protein structure; Algorithm; Peptide sequence; Biology; Genetics; Chemistry; Genome","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":[0.0007873432,0.0006107669,0.0009711874,0.0009893266,0.0002305553,0.0006178033,0.000731701,0.0004394094,0.001174739],"category_scores_gemma":[0.001731002,0.0002916074,0.0004742897,0.0007346104,0.0002403975,0.0008542817,0.0008339361,0.0008092028,0.0005279525],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004624323,"about_ca_system_score_gemma":0.0008922635,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001668719,"about_ca_topic_score_gemma":0.002670014,"domain_scores_codex":[0.9996613,0.00005587934,0.00002466367,0.00008294503,0.0001470445,0.00002816148],"domain_scores_gemma":[0.9992938,0.0002317549,0.000178649,0.00007795171,0.0001673949,0.00005039934],"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":[0.0009920283,0.0005415692,0.0568648,0.0009546268,0.0003743442,0.0004504735,0.0000967474,0.3459843,0.09647779,0.004891786,0.02230742,0.470064],"study_design_scores_gemma":[0.00006368877,0.0001077327,0.004430348,0.00001425768,0.0000207784,0.0001715011,0.00001475555,0.9659911,0.02565676,0.001310869,0.002194228,0.0000239253],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3325942,0.001926077,0.6396667,0.0004909707,0.0001075949,0.0002026188,0.005970986,0.01665706,0.002383695],"genre_scores_gemma":[0.7375062,0.0008689117,0.2512076,0.0001413059,0.00005446779,0.0001329027,0.008210463,0.0002599936,0.001618194],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.001668719,"threshold_uncertainty_score":0.004163921,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03495640096962885,"score_gpt":0.2668198993332025,"score_spread":0.2318634983635737,"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."}}