{"id":"W4382049126","doi":"10.1101/2023.06.22.546168","title":"A fast, reproducible, high-throughput variant calling workflow for evolutionary, ecological, and conservation genomics","year":2023,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Genomics and Phylogenetic Studies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":6,"is_retracted":false,"has_abstract":true,"ca_institutions":"Mount Royal University","funders":"Division of Biological Infrastructure; FAS Division of Science, Harvard University; Harvard University; National Science Foundation","keywords":"Workflow; Reuse; Computer science; Pipeline (software); Genomics; Flexibility (engineering); Data science; Population; Population genomics; Data mining; Cloud computing; Genome; Computational biology; Biology; Database; Ecology; Genetics; Gene","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.008089429,0.001236787,0.001111373,0.001995332,0.001734586,0.002743384,0.002670353,0.001495107,0.005973995],"category_scores_gemma":[0.009630705,0.001213006,0.002260298,0.00232622,0.0009453415,0.001402291,0.002814031,0.003101836,0.006426138],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00090391,"about_ca_system_score_gemma":0.003603038,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003061748,"about_ca_topic_score_gemma":0.004431573,"domain_scores_codex":[0.9952544,0.0007811143,0.0005573367,0.001563179,0.001579222,0.000264702],"domain_scores_gemma":[0.9953918,0.001160902,0.0004349459,0.001603066,0.001096634,0.0003126891],"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":[0.00158134,0.0003451872,0.0101617,0.001308073,0.0007883308,0.0008369514,0.0007350592,0.02824612,0.5267356,0.02151893,0.1064995,0.3012432],"study_design_scores_gemma":[0.0005682383,0.0004298545,0.02102549,0.0004018672,0.0003074837,0.00177371,0.0002363689,0.217658,0.4651029,0.07073043,0.22105,0.0007156194],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.02005106,0.0003604092,0.91868,0.0003961312,0.0003643458,0.0004470353,0.01414144,0.04369907,0.001860428],"genre_scores_gemma":[0.04601203,0.0002515845,0.9152529,0.0002481356,0.00008759744,0.0006733652,0.02870624,0.00558813,0.003180053],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.008089429,"threshold_uncertainty_score":0.04278147,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02426339261415852,"score_gpt":0.2280059998249233,"score_spread":0.2037426072107647,"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."}}