{"id":"W3194384536","doi":"10.1101/2021.08.13.456323","title":"Machine-learning-guided library design cycle for directed evolution of enzymes: the effects of training data composition on sequence space exploration","year":2021,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Biochemical and Structural Characterization","field":"Biochemistry, Genetics and Molecular Biology","cited_by":7,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Bio-oriented Technology Research Advancement Institution; Japan Society for the Promotion of Science; Institute of Genetics; National Agriculture and Food Research Organization; Ministry of Economy, Trade and Industry; National Institute of Advanced Industrial Science and Technology","keywords":"Directed evolution; Sequence space; Sequence (biology); Directed Molecular Evolution; Composition (language); Series (stratigraphy); Protein engineering; Function (biology); Protein sequencing; Chemical space; Computer science; Enzyme; Training (meteorology); Computational biology; Artificial intelligence; Biology; Bioinformatics; Genetics; Mathematics; Peptide sequence; Biochemistry; Gene; Geography","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":[0.0002314939,0.0002701421,0.0003126586,0.00005833486,0.00009302758,0.00005432779,0.0004525775,0.0003640314,0.000003023875],"category_scores_gemma":[0.00050558,0.0002345467,0.00009187775,0.0002020009,0.00009232705,0.0000381149,0.0003307142,0.0002465612,2.971494e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003394091,"about_ca_system_score_gemma":0.0002771921,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001298419,"about_ca_topic_score_gemma":3.209111e-7,"domain_scores_codex":[0.9983526,0.0002947231,0.0003720984,0.0006153806,0.0001782754,0.0001869041],"domain_scores_gemma":[0.9982611,0.0001071946,0.0005091428,0.0008152013,0.0002483951,0.0000589562],"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":[0.0001349433,0.00004532925,0.000189436,0.0005947032,0.00008966836,0.000001196782,0.00001796936,0.0004670392,0.9982032,0.0001827112,0.00006676606,0.00000707102],"study_design_scores_gemma":[0.0003331309,0.0001800343,0.002642754,0.0004017181,0.00007328836,3.034716e-8,0.00000518092,0.009255752,0.9867114,0.000008054333,0.0001474377,0.0002412243],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9369322,0.001317823,0.05972253,0.0001766217,0.0003619127,0.0009184747,0.0004918448,0.00007701511,0.000001590936],"genre_scores_gemma":[0.9920459,0.0003308493,0.007032008,0.00004767976,0.0002175334,0.0000923726,0.0001803256,0.0000505331,0.000002802889],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.05511371,"threshold_uncertainty_score":0.9564538,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03215822108169192,"score_gpt":0.2377377466687156,"score_spread":0.2055795255870236,"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."}}