{"id":"W4403371699","doi":"10.1101/2024.10.10.617568","title":"A Suite of Foundation Models Captures the Contextual Interplay Between Codons","year":2024,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"RNA and protein synthesis mechanisms","field":"Biochemistry, Genetics and Molecular Biology","cited_by":10,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University; Genome Canada","funders":"","keywords":"Genetic code; Codon usage bias; Synonymous substitution; Suite; Computational biology; Biology; Context (archaeology); Genetics; Computer science; Gene; Genome; Geography","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0007378338,0.001063943,0.001055077,0.0008415494,0.0005171815,0.001043341,0.001912031,0.001357386,0.002284654],"category_scores_gemma":[0.003093092,0.0004420726,0.001472567,0.0006992479,0.0008132848,0.001671445,0.0008103054,0.002101512,0.001043037],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001011066,"about_ca_system_score_gemma":0.001394483,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008060767,"about_ca_topic_score_gemma":0.01446181,"domain_scores_codex":[0.9996724,0.00009804771,0.0000187971,0.0001207729,0.00004395696,0.00004601069],"domain_scores_gemma":[0.9990366,0.000601539,0.00008159513,0.00009550634,0.0001154183,0.00006947737],"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":[0.0003278049,0.0001684305,0.008546668,0.0002237555,0.000142636,0.0003684921,0.0002425436,0.8581106,0.008910391,0.03087412,0.008346536,0.08373801],"study_design_scores_gemma":[0.0000156033,0.00003769948,0.000331775,0.00001187779,0.00001619309,0.00006730344,0.00001656601,0.9858608,0.0006218614,0.01192293,0.001084405,0.00001288931],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2635478,0.001963579,0.717543,0.001275979,0.0001909801,0.0001531047,0.004313532,0.004493644,0.00651841],"genre_scores_gemma":[0.8490496,0.001048491,0.1347669,0.0008378734,0.0001407919,0.0005039216,0.006190459,0.0004946917,0.006967246],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.008060767,"threshold_uncertainty_score":0.01602769,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01829853381787739,"score_gpt":0.2526465209186766,"score_spread":0.2343479871007992,"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."}}