{"id":"W4411224260","doi":"10.1002/pmic.13971","title":"Validation and Community Sharing of Ocean Spectral Libraries Generated by Machine Learning for Data Independent Acquisition Ocean Metaproteomic Analyses","year":2025,"lang":"en","type":"article","venue":"PROTEOMICS","topic":"Genomics and Phylogenetic Studies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"National Institute of General Medical Sciences; Simons Foundation; National Institutes of Health; National Science Foundation","keywords":"Metaproteomics; Computer science; Metagenomics; Biology","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0004397187,0.0001383735,0.000194928,0.00005272273,0.0002852668,0.00005823033,0.0003115342,0.00008773321,0.000001916836],"category_scores_gemma":[0.000116669,0.0001374119,0.0000403727,0.00007637066,0.00007031043,0.000005583007,0.0005912523,0.0001450564,9.079326e-8],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001155752,"about_ca_system_score_gemma":0.00004900238,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001626569,"about_ca_topic_score_gemma":0.00001743305,"domain_scores_codex":[0.9991733,0.0001077827,0.000244686,0.0002787981,0.00005385133,0.0001415102],"domain_scores_gemma":[0.9993451,0.00002494669,0.0001384544,0.0003932976,0.00007099834,0.00002716483],"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":[0.0001304057,0.00004801078,0.02538913,0.00007832825,0.0002400105,5.326745e-8,0.00008375513,0.0004851696,0.9729889,0.00008083149,0.000230143,0.0002452524],"study_design_scores_gemma":[0.0006075353,0.0001875675,0.002097872,0.0000121936,0.0000996002,0.000001244561,0.000123289,0.0093148,0.9858239,0.001236722,0.0003562381,0.0001390293],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9819505,0.002144382,0.01500643,0.00009000252,0.00004039857,0.0004830738,0.0002353243,0.000006011683,0.00004385685],"genre_scores_gemma":[0.9881523,0.0003951975,0.009238615,0.00005101365,0.00003880086,0.00001070968,0.001975218,0.0000148497,0.0001233391],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.02329126,"threshold_uncertainty_score":0.5603495,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04776926891472868,"score_gpt":0.3042713573600118,"score_spread":0.2565020884452831,"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."}}