{"id":"W3081097618","doi":"10.1101/2020.08.21.260109","title":"Relabeling metabolic pathway data with groups to improve prediction outcomes","year":2020,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Machine Learning in Bioinformatics","field":"Biochemistry, Genetics and Molecular Biology","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"Natural Sciences and Engineering Research Council of Canada; Genome British Columbia; Compute Canada; Genome Canada","keywords":"Computer science; Scalability; Inference; Metabolic pathway; Set (abstract data type); Artificial intelligence; Machine learning; Biology; 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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.003954658,0.001930474,0.001197896,0.002203873,0.000769705,0.001382953,0.001864176,0.001523063,0.002388429],"category_scores_gemma":[0.009909706,0.0004535135,0.001650074,0.001608315,0.0006035543,0.001981499,0.002051498,0.003176186,0.001543225],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008177158,"about_ca_system_score_gemma":0.001324289,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006118166,"about_ca_topic_score_gemma":0.007096452,"domain_scores_codex":[0.99819,0.0006393147,0.00008021047,0.0007232242,0.0002449408,0.0001224091],"domain_scores_gemma":[0.9938969,0.002679749,0.000557247,0.00171558,0.0008775011,0.0002730849],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.002414878,0.00124333,0.03857634,0.0006444113,0.0005963776,0.0003286156,0.0002972082,0.476955,0.03328627,0.002797396,0.01856707,0.4242932],"study_design_scores_gemma":[0.00004736641,0.0001853867,0.001936761,0.00003438544,0.00005175466,0.00004611582,0.000064455,0.9714381,0.01813258,0.005095066,0.002935923,0.00003203108],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.3592311,0.0015955,0.6035741,0.001019005,0.0003470908,0.0003449958,0.00550608,0.02499057,0.003391643],"genre_scores_gemma":[0.5930455,0.0002096882,0.3933636,0.0003170633,0.00008130009,0.000183924,0.009950223,0.0007736733,0.002074941],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.006118166,"threshold_uncertainty_score":0.02091449,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01351536564881809,"score_gpt":0.2320493929952831,"score_spread":0.218534027346465,"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."}}