{"id":"W3115522577","doi":"10.1093/bioinformatics/btaa842","title":"Supervised learning on phylogenetically distributed data","year":2020,"lang":"en","type":"article","venue":"Bioinformatics","topic":"Cell Image Analysis Techniques","field":"Biochemistry, Genetics and Molecular Biology","cited_by":10,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia; McGill University","funders":"Genome Canada","keywords":"Computer science; Generalization; Phylogenetic tree; Context (archaeology); Machine learning; Artificial intelligence; Independent and identically distributed random variables; Flexibility (engineering); Artificial neural network; Tree (set theory); Data mining; Biology; Mathematics; Statistics; Genetics","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":[0.000122484,0.0001362014,0.0001344748,0.00002157131,0.00005907323,0.00005021192,0.0005554548,0.00009717057,0.00003612123],"category_scores_gemma":[0.0004190596,0.0001252217,0.00005792932,0.0001276472,0.00004770846,0.000007191072,0.0004531606,0.0001207316,0.00009849],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000005745231,"about_ca_system_score_gemma":0.00003600801,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000001696591,"about_ca_topic_score_gemma":0.000001419952,"domain_scores_codex":[0.9991488,0.00002833066,0.0002667358,0.000214856,0.0001572459,0.0001839967],"domain_scores_gemma":[0.9990758,0.00001142485,0.00007445816,0.0006584884,0.0000579646,0.000121838],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0002931768,0.0002252862,0.005705547,0.000227671,0.0004100302,0.0000187817,0.0004427096,0.001173311,0.6527526,0.0001716656,0.2677403,0.07083893],"study_design_scores_gemma":[0.0006667276,0.0009553337,0.0006412069,0.00001661317,0.00008547237,0.000006107384,0.0002550212,0.1976541,0.3040449,0.0000132837,0.4951499,0.0005113775],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3641326,0.0004520845,0.5978379,0.002975117,0.00006462329,0.000903523,0.0004268669,0.0005173543,0.03268993],"genre_scores_gemma":[0.9666054,0.0002196952,0.02362102,0.002978076,0.0002047359,0.000005942153,0.006239883,0.00002612227,0.00009912871],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.6024728,"threshold_uncertainty_score":0.5106394,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02632098743923724,"score_gpt":0.2680087901917974,"score_spread":0.2416878027525601,"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."}}