{"id":"W2120688086","doi":"10.1093/bioinformatics/btq262","title":"Fast integration of heterogeneous data sources for predicting gene function with limited annotation","year":2010,"lang":"en","type":"article","venue":"Bioinformatics","topic":"Bioinformatics and Genomic Networks","field":"Biochemistry, Genetics and Molecular Biology","cited_by":139,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"Ontario Genomics; Ontario Genomics Institute; Genome Canada","keywords":"Overfitting; Computer science; Function (biology); Data mining; Annotation; Construct (python library); Source code; Artificial intelligence; Gene regulatory network; Machine learning; Association (psychology); Artificial neural network; Gene; Biology; Genetics","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":[0.0002455193,0.0001317292,0.0001218659,0.00004966526,0.00009006976,0.00004204264,0.0002352529,0.0001575259,0.000003504536],"category_scores_gemma":[0.00005935922,0.0001051403,0.00003817444,0.00007471204,0.0000617137,0.00002825777,0.00009318884,0.00009102338,0.000001861487],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000004080532,"about_ca_system_score_gemma":0.00005543502,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000004437634,"about_ca_topic_score_gemma":0.00009949333,"domain_scores_codex":[0.9991874,0.000006800906,0.0004022306,0.0001262932,0.000114367,0.0001628837],"domain_scores_gemma":[0.9988984,0.00001754437,0.0003334687,0.0005254537,0.0001756556,0.00004951092],"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":[0.001172688,0.0001655958,0.003113126,0.0004749674,0.0004299002,3.774315e-7,0.001778685,0.007778928,0.7495784,0.0004749738,0.002780156,0.2322522],"study_design_scores_gemma":[0.001643816,0.001563818,0.001161617,0.00005116799,0.0001557346,0.00005872201,0.0009292767,0.7650678,0.2157819,0.0001234171,0.01303667,0.0004261099],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5899263,0.00004020767,0.4090614,0.00002059788,0.0002023511,0.0003197382,0.0001960122,0.00001484157,0.0002184998],"genre_scores_gemma":[0.914431,0.00002320047,0.08094632,0.0001088967,0.0002518708,0.00001705235,0.004155602,0.0000170772,0.000048963],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.7572888,"threshold_uncertainty_score":0.4287498,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01543798048489488,"score_gpt":0.2300689522546752,"score_spread":0.2146309717697803,"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."}}