{"id":"W3134927026","doi":"10.1101/2021.03.05.434175","title":"Evotuning protocols for Transformer-based variant effect prediction on multi-domain proteins","year":2021,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Genomics and Phylogenetic Studies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Japan Society for the Promotion of Science; Ministry of Education, Culture, Sports, Science and Technology; National Institute of Advanced Industrial Science and Technology; Institute of Genetics; Japan Agency for Medical Research and Development","keywords":"Computer science; Transformer; Machine learning; Artificial intelligence; Feature learning; Artificial neural network; Feature engineering; Data mining; Deep learning; Engineering; Voltage","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001812246,0.0008146064,0.0005615309,0.0007218251,0.0005620844,0.0006645876,0.001359295,0.0007058121,0.004702785],"category_scores_gemma":[0.005903497,0.0004457299,0.0006302843,0.0006120133,0.0005244396,0.001228027,0.002547945,0.001491919,0.001718501],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006616026,"about_ca_system_score_gemma":0.0008478214,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001004769,"about_ca_topic_score_gemma":0.002044979,"domain_scores_codex":[0.9990119,0.0002087306,0.0001006877,0.0003091214,0.0002598103,0.000109815],"domain_scores_gemma":[0.9984192,0.0007202004,0.00008558544,0.0004727232,0.0002230528,0.00007934269],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001460807,0.0003215848,0.005681974,0.0008673372,0.0002068334,0.0008298311,0.0007667865,0.08896083,0.3859776,0.02603862,0.02010049,0.4687874],"study_design_scores_gemma":[0.00009856987,0.0001678546,0.002065346,0.00005095142,0.00006756955,0.0004130436,0.0001664691,0.6511967,0.3030114,0.02684662,0.01582422,0.0000912863],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.07477557,0.0003721635,0.89752,0.0001776993,0.0000902122,0.0002788495,0.0009053985,0.02346936,0.002410837],"genre_scores_gemma":[0.4611471,0.0003251527,0.5290857,0.0002019917,0.00003060367,0.0005574187,0.003280271,0.002734533,0.00263728],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.004702785,"threshold_uncertainty_score":0.01573235,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01661938041256742,"score_gpt":0.2443029632778607,"score_spread":0.2276835828652933,"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."}}