{"id":"W3004898987","doi":"10.1021/acs.jproteome.9b00555","title":"Repeat-Preserving Decoy Database for False Discovery Rate Estimation in Peptide Identification","year":2020,"lang":"en","type":"article","venue":"Journal of Proteome Research","topic":"Advanced Proteomics Techniques and Applications","field":"Chemistry","cited_by":44,"is_retracted":false,"has_abstract":true,"ca_institutions":"SickKids Foundation; Hospital for Sick Children; University of Toronto; University of Waterloo","funders":"Natural Sciences and Engineering Research Council of Canada; Canada Research Chairs; Canada Foundation for Innovation; Ontario Genomics; Genome Canada","keywords":"Decoy; False discovery rate; De Bruijn sequence; Computer science; Identification (biology); Sequence database; Database search engine; Mascot; Data mining; Database; Mathematics; Biology; Information retrieval; Search engine","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":[0.002396429,0.0001001391,0.0002000783,0.0001820053,0.0001429955,0.000161904,0.0005442968,0.00007642732,0.00004858709],"category_scores_gemma":[0.002793046,0.00008491188,0.00008712144,0.000412303,0.00005375727,0.0009512126,0.0001758939,0.000680519,0.000008058659],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001830016,"about_ca_system_score_gemma":0.0001738138,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001979406,"about_ca_topic_score_gemma":0.000005949801,"domain_scores_codex":[0.9982655,0.00007884914,0.0007024185,0.0002507967,0.0004187019,0.0002837316],"domain_scores_gemma":[0.9983283,0.0003074996,0.0003840028,0.0003503943,0.0004930703,0.0001367744],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.000290059,0.00007402215,0.0003588842,0.0004095674,0.0000122836,0.00001198156,0.0001590553,0.0008924813,0.9938358,0.0003825635,0.001864531,0.001708787],"study_design_scores_gemma":[0.0008966238,0.0001169904,0.0002705069,0.0003941215,0.0000115624,0.00001728246,0.0003684527,0.04161824,0.9353501,0.01542669,0.005354366,0.000175074],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4133087,0.0001474827,0.578961,0.006138557,0.00001646687,0.001052697,0.00008438005,0.00003122176,0.0002595532],"genre_scores_gemma":[0.8201454,0.0002266514,0.1776422,0.00004166952,0.0002852686,0.0006516111,0.00006017583,0.00004297894,0.0009041167],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.4068367,"threshold_uncertainty_score":0.3462606,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1032918760521459,"score_gpt":0.4207004083951112,"score_spread":0.3174085323429653,"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."}}