{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.000169775,0.00004794456,0.00004369654,0.00004894064,0.00004777721,0.000008314617,0.00009714484,0.00004926751,0.000001015957],"category_scores_gemma":[0.00006459639,0.00003174006,0.00002941488,0.0003078412,0.00002407627,0.00000177752,0.000008536899,0.00004568681,4.215765e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000005797754,"about_ca_system_score_gemma":0.00005906304,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002526086,"about_ca_topic_score_gemma":0.0001406245,"domain_scores_codex":[0.999648,0.00001080693,0.0001288461,0.00007822581,0.00004769868,0.00008639984],"domain_scores_gemma":[0.9997531,0.00001311469,0.00004406967,0.0001318301,0.00005267329,0.000005266314],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.00005502668,0.00001895767,0.00004629531,0.0000254632,0.000009996842,7.652255e-7,0.0009655825,0.01398315,0.9797301,0.000575867,0.00006067403,0.004528077],"study_design_scores_gemma":[0.001272638,0.0001455049,0.002775203,0.00003467788,0.00002506308,0.000002976955,0.003047154,0.2063607,0.7845297,0.001019512,0.0005892668,0.0001976137],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9853905,0.00009329247,0.01353704,0.0004301151,0.00002277318,0.0003976394,0.00008560828,0.000007749556,0.00003528043],"genre_scores_gemma":[0.9987486,0.000003367264,0.0003859158,0.0002185442,0.00002926877,0.0001404133,0.0003772842,0.00001011088,0.00008649949],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.1952004,"threshold_uncertainty_score":0.1294322,"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."}}