{"id":"W4317566064","doi":"10.1101/2023.01.18.524644","title":"Functional domain annotation by structural similarity","year":2023,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Genomics and Phylogenetic Studies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"Canadian Institutes of Health Research","keywords":"Annotation; Computational biology; Structural similarity; Structural alignment; Protein domain; In silico; Domain (mathematical analysis); Similarity (geometry); Sequence alignment; Proteome; UniProt; Computer science; Benchmark (surveying); Sequence (biology); Protein sequencing; Biology; Bioinformatics; Peptide sequence; Genetics; 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":[0.001208417,0.0009532397,0.0008578983,0.005954201,0.0005951177,0.001206378,0.0008876451,0.0006661672,0.008540341],"category_scores_gemma":[0.003326405,0.000270358,0.0008503226,0.002985818,0.0003275362,0.001233463,0.0009062915,0.0006013637,0.003968611],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004392166,"about_ca_system_score_gemma":0.0005106415,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0007722997,"about_ca_topic_score_gemma":0.0007762275,"domain_scores_codex":[0.9988652,0.0002887058,0.0001451189,0.0002887254,0.0003000532,0.0001120907],"domain_scores_gemma":[0.998107,0.0007117396,0.0002568451,0.0003350454,0.0004815492,0.0001078664],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.002769503,0.000554799,0.04415195,0.00298652,0.0003573437,0.0008422155,0.0005162875,0.01252929,0.5562574,0.007999861,0.01421962,0.3568152],"study_design_scores_gemma":[0.0002497382,0.0009329822,0.07669017,0.0004023524,0.0003628959,0.003471978,0.0007747755,0.3502765,0.4502936,0.02311644,0.09321024,0.0002183278],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"methods","genre_scores_codex":[0.5459036,0.002870107,0.3939894,0.0004228632,0.000203895,0.0005377361,0.02227676,0.01964996,0.0141456],"genre_scores_gemma":[0.7203837,0.0006176982,0.2405019,0.0001048315,0.0000570612,0.0002366096,0.03394192,0.0008990061,0.003257385],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.008540341,"threshold_uncertainty_score":0.02857023,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01551278335407071,"score_gpt":0.2195798022130329,"score_spread":0.2040670188589622,"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."}}