{"id":"W3206605025","doi":"10.1016/j.xpro.2021.100888","title":"Protocol for using NoBadWordsCombiner to merge and minimize “bad words” from BLAST hits against multiple eukaryotic gene annotation databases","year":2021,"lang":"en","type":"article","venue":"STAR Protocols","topic":"Genomics and Phylogenetic Studies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"Western University; Dalhousie University","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Computer science; Merge (version control); Annotation; Unix; Protocol (science); Gene nomenclature; Database; Coding (social sciences); Information retrieval; Programming language; Artificial intelligence; Software; Biology; Medicine","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.000121331,0.0002434408,0.0002531038,0.0000340119,0.0001637908,0.00005081019,0.0001214047,0.0000761548,0.00001268307],"category_scores_gemma":[0.0002197795,0.0002362176,0.00007387503,0.0001016243,0.00004596897,0.000003315899,0.0002754387,0.00005005686,0.0000024653],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001785563,"about_ca_system_score_gemma":0.000142793,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003691782,"about_ca_topic_score_gemma":0.0001712734,"domain_scores_codex":[0.9985756,0.00007041584,0.0003200406,0.0006076721,0.0001174759,0.0003087946],"domain_scores_gemma":[0.9991058,0.0000458151,0.0001125446,0.0004058109,0.0002087887,0.0001212355],"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":[0.000536678,0.0001383043,0.01102966,0.00009930741,0.0001044829,0.000004660969,0.0001161129,0.0002824185,0.9845143,0.00001444297,0.0006994822,0.002460128],"study_design_scores_gemma":[0.00556375,0.0004017397,0.01712304,0.0001835243,0.00003630548,0.000004451055,0.0001981737,0.001615121,0.8758935,0.0001267825,0.0982665,0.0005871479],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"protocol","genre_scores_codex":[0.537409,0.00003620856,0.005719876,0.00008325735,0.00005262962,0.4558407,0.0008049952,0.000007157562,0.0000461252],"genre_scores_gemma":[0.03908518,0.000002777433,0.08063301,0.0003436253,0.0003046872,0.8791218,0.0003499316,0.00004627493,0.0001127353],"genre_candidate":"protocol","genre_consensus":null,"teacher_disagreement_score":0.4983238,"threshold_uncertainty_score":0.9632672,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06534366516059398,"score_gpt":0.3465030867927246,"score_spread":0.2811594216321306,"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."}}