{"id":"W1595684610","doi":"10.1186/1471-2105-5-61","title":"An approach to large scale identification of non-obvious structural similarities between proteins","year":2004,"lang":"en","type":"article","venue":"BMC Bioinformatics","topic":"Machine Learning in Bioinformatics","field":"Biochemistry, Genetics and Molecular Biology","cited_by":7,"is_retracted":false,"has_abstract":true,"ca_institutions":"BC Cancer Agency; University of British Columbia","funders":"Genome Prairie; Genome British Columbia; Michael Smith Health Research BC","keywords":"Threading (protein sequence); Computational biology; Biology; Genome; Sequence alignment; Genetics; Structural similarity; Peptide sequence; Protein structure; Sequence (biology); Protein sequencing; Gene; DNA microarray; Virulence; Gene expression","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":[0.001610442,0.0006502037,0.0007049809,0.001817825,0.0005405525,0.0009380566,0.0009170792,0.0006739865,0.001464818],"category_scores_gemma":[0.004019956,0.0003676046,0.0008158315,0.001608602,0.0007146642,0.001043025,0.001439092,0.001197149,0.0006706975],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005620297,"about_ca_system_score_gemma":0.0008485876,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0004343733,"about_ca_topic_score_gemma":0.0006041621,"domain_scores_codex":[0.9987848,0.0002527975,0.0001035456,0.0003457842,0.0004477279,0.00006529073],"domain_scores_gemma":[0.9980255,0.0007789508,0.0003631135,0.0003242066,0.0003694023,0.0001388249],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0009321552,0.0004978744,0.01086293,0.0008224747,0.0003166041,0.0005587837,0.0003669093,0.02508458,0.5320612,0.01242485,0.00238132,0.4136902],"study_design_scores_gemma":[0.0001532077,0.0006004306,0.01161884,0.00006035655,0.0001377856,0.002168682,0.0001398119,0.752207,0.185414,0.03654497,0.01087891,0.00007599263],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.04626504,0.000183989,0.9502268,0.0000754154,0.00001915848,0.000105436,0.000288228,0.002325745,0.0005102102],"genre_scores_gemma":[0.1778013,0.0000840091,0.8202831,0.00005999755,0.00001574399,0.0001761018,0.000870787,0.0001427056,0.0005663418],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.001817825,"threshold_uncertainty_score":0.008516967,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.009320389599279316,"score_gpt":0.2698542294167605,"score_spread":0.2605338398174811,"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."}}