{"id":"W4416257247","doi":"10.1371/journal.pone.0335863","title":"Improved CRISPR/Cas9 off-target prediction with DNABERT and epigenetic features","year":2025,"lang":"en","type":"article","venue":"PLoS ONE","topic":"CRISPR and Genetic Engineering","field":"Biochemistry, Genetics and Molecular Biology","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Institute of Genetics; University of Tokyo","keywords":"Interpretability; Leverage (statistics); Epigenetics; Deep learning; Epigenome; Genomics; Genome; Benchmark (surveying)","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":[0.0006505738,0.001161423,0.0008162265,0.0004425301,0.000234597,0.0006490363,0.0009740986,0.0009466778,0.0008273952],"category_scores_gemma":[0.001575768,0.0002832242,0.0006094748,0.0003083927,0.0003175448,0.0006497487,0.0005862204,0.001365324,0.0003333268],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006868908,"about_ca_system_score_gemma":0.000969038,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008023303,"about_ca_topic_score_gemma":0.01215122,"domain_scores_codex":[0.9997577,0.00004189239,0.00001183688,0.0000876528,0.00006634166,0.00003451973],"domain_scores_gemma":[0.999486,0.0003041257,0.0000473507,0.0000532963,0.00007770053,0.00003165296],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0002917515,0.0001732148,0.006698316,0.0001587247,0.0001399415,0.0002463878,0.00004246306,0.8592029,0.0307255,0.001738107,0.00331641,0.09726628],"study_design_scores_gemma":[0.000006697428,0.00004659941,0.0005618799,0.000004127115,0.00001230251,0.0000258698,0.00000425937,0.9898448,0.008383864,0.0007221729,0.0003762695,0.0000112582],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4539911,0.00216102,0.5223626,0.000792011,0.0001798945,0.0001172538,0.00197293,0.01264372,0.005779518],"genre_scores_gemma":[0.9132024,0.0004020923,0.07913643,0.0004594835,0.00002765386,0.0000870045,0.002555079,0.0002762225,0.003853618],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.008023303,"threshold_uncertainty_score":0.01595318,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0055633272442177,"score_gpt":0.2295917718844263,"score_spread":0.2240284446402086,"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."}}