{"id":"W4386278355","doi":"10.1111/1462-2920.16487","title":"Data‐driven identification of major axes of functional variation in bacteria","year":2023,"lang":"en","type":"article","venue":"Environmental Microbiology","topic":"Genomics and Phylogenetic Studies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université du Québec à Montréal","funders":"Natural Sciences and Engineering Research Council of Canada; Canada Research Chairs","keywords":"Biology; Metagenomics; Evolutionary biology; Phylogenetic tree; Trait; Phylogenetics; Ecology; Ecosystem; Functional diversity; Biodiversity; Taxonomic rank; Habitat; Variation (astronomy); Bacterial genome size; Phylogenetic diversity; Genome; Taxon; Genetics; Gene","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":[0.0001229252,0.00006191561,0.00009843915,0.00004365468,0.00001583573,0.000001200796,0.0001312194,0.00006986873,0.00003177083],"category_scores_gemma":[0.0000116877,0.00006414884,0.00002341492,0.00004198897,0.00009317775,9.383011e-7,0.0002297142,0.00002196187,0.00001687689],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000008097567,"about_ca_system_score_gemma":0.000009054781,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000008865207,"about_ca_topic_score_gemma":0.00001968561,"domain_scores_codex":[0.9994052,0.00003784705,0.000235777,0.0002133525,0.00001978997,0.00008803828],"domain_scores_gemma":[0.9996232,0.00001056717,0.0001109601,0.0002423626,0.000003900921,0.000009017976],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"observational","study_design_scores_codex":[0.00002867184,0.00004152159,0.01555716,0.000006628431,0.00003683302,1.634199e-7,0.00004230214,0.0001911093,0.9837189,0.00001261066,0.0001968484,0.0001672428],"study_design_scores_gemma":[0.0002727801,0.00006058203,0.655512,0.000001914885,0.000008769713,0.000003098829,0.00005135686,0.00005284775,0.3418214,0.00003648939,0.00212306,0.00005573606],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9983878,0.0002145904,0.0001007592,0.00003434338,0.0001458792,0.00008797669,0.001013685,0.000001074695,0.0000139282],"genre_scores_gemma":[0.9969831,0.0002918517,0.00017249,0.00001273285,0.00003498895,0.00000655132,0.002392967,0.000006478107,0.00009890704],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.6418975,"threshold_uncertainty_score":0.2615914,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01212186522745259,"score_gpt":0.2103770489810347,"score_spread":0.1982551837535821,"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."}}