{"id":"W2114857957","doi":"10.1186/1471-2105-4-16","title":"Domain fusion analysis by applying relational algebra to protein sequence and domain databases","year":2003,"lang":"en","type":"article","venue":"BMC Bioinformatics","topic":"Bioinformatics and Genomic Networks","field":"Biochemistry, Genetics and Molecular Biology","cited_by":23,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"National Cancer Institute; Canadian Institutes of Health Research","keywords":"Computer science; Domain (mathematical analysis); UniProt; Protein domain; Sequence database; Hidden Markov model; Computational biology; Relational database; Database; Protein structure database; Sequence (biology); Protein sequencing; Data mining; Artificial intelligence; Biology; Peptide sequence; Gene; Genetics; Mathematics","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":[0.0057404,0.0007400996,0.0008499977,0.003888466,0.001057499,0.00387554,0.002070914,0.0004847269,0.003387638],"category_scores_gemma":[0.01328438,0.0005330591,0.002416432,0.003918122,0.001132065,0.005087101,0.002889258,0.001205607,0.001627619],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001178797,"about_ca_system_score_gemma":0.002024071,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004718405,"about_ca_topic_score_gemma":0.003838231,"domain_scores_codex":[0.9952355,0.001319594,0.0007446511,0.0008100386,0.001730572,0.0001596256],"domain_scores_gemma":[0.9937796,0.003368903,0.0004674484,0.001436902,0.0008015482,0.000145602],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0005790573,0.0003200978,0.008022186,0.0008363432,0.0005353013,0.000930018,0.001019355,0.1047941,0.01934672,0.3578461,0.009947794,0.4958229],"study_design_scores_gemma":[0.0001071818,0.0001672766,0.001010594,0.0001300113,0.0002351998,0.0007603148,0.0004538324,0.6403333,0.0270406,0.2974682,0.03219797,0.00009554959],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.004800751,0.0001106617,0.9906308,0.0001145034,0.00001375983,0.00009178687,0.0003747196,0.003016574,0.0008464638],"genre_scores_gemma":[0.06972768,0.0002786267,0.9272323,0.00006910136,0.00003218725,0.0001579137,0.001790199,0.000215191,0.0004969256],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.0057404,"threshold_uncertainty_score":0.03035849,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01848185516848864,"score_gpt":0.2449777261008124,"score_spread":0.2264958709323238,"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."}}