{"id":"W2118584916","doi":"10.1093/bioinformatics/btp552","title":"Reproducing the manual annotation of multiple sequence alignments using a SVM classifier","year":2009,"lang":"en","type":"article","venue":"Bioinformatics","topic":"Genomics and Phylogenetic Studies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":10,"is_retracted":false,"has_abstract":true,"ca_institutions":"Dalhousie University","funders":"","keywords":"Computer science; Annotation; Classifier (UML); Support vector machine; Software; Artificial intelligence; Data mining; Sequence (biology); Machine learning; Multiple sequence alignment; Pattern recognition (psychology); Sequence alignment; Programming language; Peptide sequence","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0001743081,0.00008391252,0.00008379473,0.00001773332,0.0000865605,0.00001260326,0.0001179679,0.00004706885,0.000001047045],"category_scores_gemma":[0.00007457557,0.00006161709,0.00004300397,0.00005483847,0.00005121755,0.000001883416,0.00004974176,0.00003052592,0.000001747344],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000009471119,"about_ca_system_score_gemma":0.00003171169,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000009570968,"about_ca_topic_score_gemma":0.000002986688,"domain_scores_codex":[0.9994061,0.00001424169,0.0002518834,0.0001025567,0.00009777274,0.0001274295],"domain_scores_gemma":[0.9994522,0.000009005249,0.0001503797,0.0003012996,0.00006649581,0.00002058178],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.00002716736,0.00002734432,0.003179639,0.00002287591,0.00004450769,3.048456e-7,0.001137188,0.001401121,0.9824789,0.00009443922,0.0004206689,0.01116588],"study_design_scores_gemma":[0.001358504,0.0008270676,0.0826443,0.00007744096,0.0001082034,0.00005213344,0.003736861,0.07169653,0.8213296,0.0006793361,0.01686102,0.000629022],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.996305,0.0002045056,0.002400953,0.00008584893,0.00008811554,0.0001531628,0.00002027314,0.000002077446,0.0007400335],"genre_scores_gemma":[0.9875086,0.00007984317,0.01202947,0.0002327903,0.00006826596,0.000002025763,0.00001649406,0.000004204527,0.00005833697],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.1611493,"threshold_uncertainty_score":0.2512672,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04061509860338373,"score_gpt":0.2885442176537646,"score_spread":0.2479291190503808,"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."}}