{"id":"W2005813445","doi":"10.1093/database/baq031","title":"LPS-annotate: complete annotation of compositionally biased regions in the protein knowledgebase","year":2011,"lang":"en","type":"article","venue":"Database","topic":"Genomics and Phylogenetic Studies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":24,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"McGill University","keywords":"UniProt; Annotation; Protein sequencing; Computational biology; Genome; Biology; Database; Computer science; Gene; Genetics; Peptide sequence","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":true,"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":[0.00253512,0.002615144,0.001759391,0.005949377,0.001423802,0.003615297,0.002664575,0.002259407,0.01264554],"category_scores_gemma":[0.009431352,0.001316701,0.001481962,0.005388313,0.0006366582,0.00389052,0.003946348,0.001981434,0.01682484],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001205379,"about_ca_system_score_gemma":0.003062412,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002881417,"about_ca_topic_score_gemma":0.003919978,"domain_scores_codex":[0.9983268,0.0002073149,0.0002688391,0.0005061761,0.0005441493,0.0001466544],"domain_scores_gemma":[0.9964574,0.001140635,0.0005936486,0.0008840511,0.0006827396,0.0002414],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.003234102,0.0006425108,0.01909129,0.01050948,0.0006751618,0.001817641,0.00259705,0.01260831,0.1263448,0.01359177,0.4488811,0.3600068],"study_design_scores_gemma":[0.0004619686,0.0003470467,0.0225528,0.001207199,0.0004099221,0.002093526,0.001004828,0.05834441,0.08701586,0.02229316,0.8037851,0.0004841082],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"dataset","genre_gemma":"methods","genre_scores_codex":[0.02795052,0.002720609,0.33717,0.0005425073,0.0002421594,0.0005542537,0.4029103,0.2188161,0.009093562],"genre_scores_gemma":[0.05376603,0.001688535,0.357296,0.000403969,0.00007429885,0.001037934,0.565402,0.0175098,0.002821363],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.01264554,"threshold_uncertainty_score":0.04230356,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06237460809254645,"score_gpt":0.260019375891962,"score_spread":0.1976447677994155,"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."}}