{"id":"W4393895133","doi":"10.5281/zenodo.8101702","title":"MetaPep: A core peptide database for faster human gut metaproteomics database searches","year":2023,"lang":"en","type":"dataset","venue":"Zenodo (CERN European Organization for Nuclear Research)","topic":"Machine Learning in Bioinformatics","field":"Biochemistry, Genetics and Molecular Biology","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Ottawa","funders":"","keywords":"Metaproteomics; Database; Computer science; Core (optical fiber); Computational biology; Chemistry; Biology; Proteomics; Biochemistry","routes":{"ca_aff":true,"ca_fund":false,"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":[0.001903423,0.00265439,0.00141919,0.003694467,0.0009757985,0.001458579,0.003283579,0.001842258,0.01097044],"category_scores_gemma":[0.003258927,0.0006123465,0.001462753,0.003403323,0.0003867646,0.001183915,0.002573458,0.001674793,0.01591334],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009339619,"about_ca_system_score_gemma":0.001769872,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007141795,"about_ca_topic_score_gemma":0.01322462,"domain_scores_codex":[0.9990059,0.0001580624,0.0001246311,0.0003549235,0.0002238966,0.0001326528],"domain_scores_gemma":[0.9993001,0.0001519317,0.00008156172,0.0002078392,0.0001460886,0.0001124857],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.00171488,0.000429466,0.01173001,0.003263034,0.0005534478,0.0006558098,0.0001469393,0.003622792,0.0105675,0.002436998,0.9220024,0.0428766],"study_design_scores_gemma":[0.001908975,0.000326828,0.02235483,0.0004816189,0.0003475779,0.001319737,0.0003058721,0.0160632,0.01475714,0.006503357,0.9354516,0.0001792695],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.004910689,0.0005120983,0.001943846,0.000124233,0.00006237796,0.0001045348,0.9889116,0.002533965,0.0008966489],"genre_scores_gemma":[0.002480917,0.0001241383,0.004371737,0.00004456554,0.00000608639,0.0001873574,0.9924464,0.00008000993,0.0002588002],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.01097044,"threshold_uncertainty_score":0.03669977,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06709769633482267,"score_gpt":0.3102441220615064,"score_spread":0.2431464257266838,"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."}}