{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0005504126,0.0002151762,0.0002098435,0.0001169942,0.0002006766,0.00006599651,0.0001465388,0.000139173,0.00003267426],"category_scores_gemma":[0.00005621259,0.0001971989,0.00008810125,0.0003670542,0.00007827564,0.00002496845,0.0001542387,0.0001085482,0.00002678524],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002318217,"about_ca_system_score_gemma":0.00008486524,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000008557455,"about_ca_topic_score_gemma":0.00004430599,"domain_scores_codex":[0.9986978,0.00004723127,0.0005062342,0.0002128813,0.0002163663,0.0003194541],"domain_scores_gemma":[0.9991238,0.00002189321,0.0001812731,0.0004259988,0.00005483555,0.0001921928],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0008142187,0.0006412505,0.04928452,0.002046985,0.003091687,0.00001202695,0.006563339,0.01950537,0.4554582,0.3295282,0.08863075,0.04442343],"study_design_scores_gemma":[0.002415591,0.0004676988,0.001148619,0.0001405006,0.0003150807,0.00008726039,0.003205141,0.0513131,0.01635942,0.004453251,0.9181909,0.001903431],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.2147928,0.0002740301,0.7816766,0.00005497875,0.00004156262,0.0006083878,0.000179633,0.00001599877,0.002356019],"genre_scores_gemma":[0.08800323,0.00006552679,0.9091431,0.0007692515,0.00006192128,0.0001120045,0.001392165,0.00002176048,0.0004309696],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.8295602,"threshold_uncertainty_score":0.8041536,"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."}}