{"id":"W4392844094","doi":"10.1093/nargab/lqae005","title":"Functional domain annotation by structural similarity","year":2024,"lang":"en","type":"article","venue":"NAR Genomics and Bioinformatics","topic":"Genomics and Phylogenetic Studies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":19,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"Canadian Institutes of Health Research","keywords":"Annotation; Computational biology; Structural similarity; In silico; Protein domain; Domain (mathematical analysis); Similarity (geometry); Structural alignment; Sequence alignment; Proteome; UniProt; Biology; Sequence (biology); Protein sequencing; Benchmark (surveying); Computer science; Bioinformatics; Genetics; Peptide sequence; Artificial intelligence; Geography","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":[0.001246661,0.0008896518,0.0008472149,0.006222958,0.0005914202,0.001105403,0.0008718146,0.0006089984,0.008372777],"category_scores_gemma":[0.003549899,0.0002593179,0.0009425549,0.002996022,0.0003067155,0.001242818,0.0008858124,0.0005596699,0.003463812],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004392291,"about_ca_system_score_gemma":0.0004960991,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0008646972,"about_ca_topic_score_gemma":0.000907174,"domain_scores_codex":[0.9988798,0.0002917152,0.0001599802,0.000267231,0.0002921843,0.0001090061],"domain_scores_gemma":[0.9979659,0.0007853809,0.000286793,0.0003087825,0.0005451075,0.0001079882],"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":[0.002649112,0.0005647654,0.04861657,0.002986327,0.0003916032,0.0008653697,0.0004952931,0.01102564,0.5711436,0.005918914,0.01026814,0.3450747],"study_design_scores_gemma":[0.0002589661,0.001206101,0.1004906,0.0004738953,0.0004678122,0.004343539,0.000868884,0.3572037,0.4295518,0.01804806,0.08683831,0.0002483786],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"methods","genre_scores_codex":[0.5840777,0.002904861,0.3640577,0.0003672148,0.0001697321,0.0005996197,0.0186364,0.01594919,0.01323757],"genre_scores_gemma":[0.7405499,0.000587378,0.2291354,0.00009899146,0.00004664459,0.0002428797,0.02606397,0.0006376223,0.002637304],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.008372777,"threshold_uncertainty_score":0.02800977,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.007423367596102924,"score_gpt":0.2145816978091313,"score_spread":0.2071583302130283,"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."}}