{"id":"W2938712064","doi":"10.1093/bioinformatics/btz266","title":"Discovering novel mutation signatures by latent Dirichlet allocation with variational Bayes inference","year":2019,"lang":"en","type":"article","venue":"Bioinformatics","topic":"Cancer Genomics and Diagnostics","field":"Biochemistry, Genetics and Molecular Biology","cited_by":13,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Core Research for Evolutional Science and Technology; Institute of Genetics; Japan Society for the Promotion of Science; Waseda University; Ministry of Education, Culture, Sports, Science and Technology","keywords":"Mutation; Latent Dirichlet allocation; Computer science; Cluster analysis; Bayes' theorem; Inference; Mutation rate; Genetics; Computational biology; Biology; Bayesian probability; Artificial intelligence; Topic model; Gene","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.00006016843,0.0001077424,0.00007758538,0.00002545997,0.00003958853,0.00006139619,0.00009626321,0.00007843607,0.0000151926],"category_scores_gemma":[0.00004851762,0.00009153063,0.00002404086,0.00005870294,0.0000199285,0.00001689792,0.00004919143,0.00005019097,0.0000197327],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000242888,"about_ca_system_score_gemma":0.00009914402,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002234566,"about_ca_topic_score_gemma":0.00002153759,"domain_scores_codex":[0.9994354,0.000004036834,0.000176924,0.0001140542,0.0001416354,0.0001279239],"domain_scores_gemma":[0.9995683,0.0000275727,0.0001127325,0.0001684802,0.00008238811,0.00004057239],"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":[0.0004533086,0.0004598462,0.03992584,0.0004433288,0.000413611,0.000001765852,0.001820531,0.2553051,0.6619524,0.01452323,0.01494861,0.009752339],"study_design_scores_gemma":[0.008500147,0.003226186,0.09415293,0.0002909984,0.0002024885,0.00005247977,0.001132817,0.5047256,0.2571709,0.0004930258,0.1273242,0.002728248],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7424332,0.0001791983,0.2552865,0.0001383126,0.000180061,0.0002997258,0.0001974736,0.00001457322,0.001270956],"genre_scores_gemma":[0.987174,0.00009076248,0.01090888,0.0003432888,0.00006092285,0.00001328214,0.00122292,0.00001139685,0.0001744884],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4047816,"threshold_uncertainty_score":0.3732511,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.005256585999953559,"score_gpt":0.2151006418825204,"score_spread":0.2098440558825668,"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."}}