{"id":"W4385324506","doi":"10.1016/j.xinn.2023.100487","title":"Deciphering “the language of nature”: A transformer-based language model for deleterious mutations in proteins","year":2023,"lang":"en","type":"article","venue":"The Innovation","topic":"Genomics and Rare Diseases","field":"Biochemistry, Genetics and Molecular Biology","cited_by":35,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Fundamental Research Funds for the Central Universities; Google; Sun Yat-sen University; Cancer Research Institute; CHEO Research Institute; National Institutes of Health","keywords":"Transformer; Computer science; Linguistics; Computational biology; Biology; Engineering; Philosophy; Electrical engineering","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":[0.0004408447,0.0005153102,0.0003224559,0.0005427064,0.0001820973,0.0004892027,0.0008125052,0.0005975825,0.0009950214],"category_scores_gemma":[0.001153346,0.0002162049,0.0007616092,0.0002711763,0.0004984225,0.001146737,0.0004990639,0.0009386135,0.0003851549],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006180418,"about_ca_system_score_gemma":0.0008542292,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004683094,"about_ca_topic_score_gemma":0.005255144,"domain_scores_codex":[0.9998944,0.0000278,0.000006674692,0.00003414613,0.00001953351,0.0000174479],"domain_scores_gemma":[0.9996502,0.0002020077,0.00004164865,0.00002432752,0.00005338864,0.00002841914],"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":[0.0004190429,0.0001721949,0.004610378,0.0001364673,0.0001084438,0.0005349555,0.0003227001,0.7832431,0.04185018,0.04330741,0.005420059,0.1198751],"study_design_scores_gemma":[0.00000430949,0.00001449809,0.0001034793,0.00000202681,0.000005886813,0.00002977617,0.000005793127,0.9926434,0.00102048,0.005917054,0.0002492917,0.000003997373],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.1778869,0.0007823732,0.8139935,0.001330591,0.00011376,0.00005041861,0.0007377585,0.002227854,0.00287681],"genre_scores_gemma":[0.926972,0.0005486091,0.0674079,0.000458535,0.00005540329,0.00006239154,0.0007654946,0.0001491708,0.003580546],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.004683094,"threshold_uncertainty_score":0.009311676,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01217872581937492,"score_gpt":0.2865710491205445,"score_spread":0.2743923233011696,"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."}}