{"id":"W2985622460","doi":"10.1074/mcp.tir119.001752","title":"Assessing Protein Sequence Database Suitability Using De Novo Sequencing","year":2019,"lang":"en","type":"article","venue":"Molecular & Cellular Proteomics","topic":"Genomics and Phylogenetic Studies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":43,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Office of Extramural Research, National Institutes of Health; National Institute of General Medical Sciences","keywords":"Computational biology; Sequence (biology); Sequence database; Protein sequencing; Computer science; Biology; Database; Peptide sequence; Genetics; Gene","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0007332438,0.0003413934,0.0002859123,0.00005669894,0.000166894,0.0001368218,0.0003939234,0.0002171152,0.00001832391],"category_scores_gemma":[0.0001306035,0.0003718694,0.0001683081,0.0001508689,0.000119388,0.00000812554,0.0004292375,0.0002302882,0.00001672778],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001902227,"about_ca_system_score_gemma":0.0005576862,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001201847,"about_ca_topic_score_gemma":0.000009231308,"domain_scores_codex":[0.9977845,0.0001978299,0.0003647455,0.0007892616,0.0002298776,0.000633804],"domain_scores_gemma":[0.9985517,0.000009247335,0.0001726489,0.0009723404,0.0001487542,0.0001452622],"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":[0.00002730146,0.00003498093,0.004622236,0.0001017487,0.00006189584,0.00004394586,0.00005783874,0.00198453,0.9927482,0.0001513262,0.00000185748,0.0001641156],"study_design_scores_gemma":[0.0004499213,0.0001036791,0.0001779828,0.00005336475,0.00003367835,0.00005920463,0.0001543587,0.003166563,0.9945848,0.0003464459,0.0004169718,0.0004530113],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9525296,0.0007807899,0.04491358,0.00004254733,0.0001206466,0.001216654,0.00003390964,0.00001360495,0.0003486487],"genre_scores_gemma":[0.9259381,0.00001641053,0.07346741,0.0001811839,0.0001052734,0.00005693504,0.00007623022,0.00006993031,0.00008847132],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.02855383,"threshold_uncertainty_score":0.9998733,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03392472601878817,"score_gpt":0.2786965261732743,"score_spread":0.2447718001544862,"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."}}