{"id":"W4409568964","doi":"10.1186/s12864-025-11569-y","title":"Kastor: a reference-based comparative approach for assessment and correction of gene-fragmenting errors in long-read assemblies of small genomes","year":2025,"lang":"en","type":"article","venue":"BMC Genomics","topic":"Genomics and Phylogenetic Studies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Pseudogene; Genome; Biology; Computational biology; Gene; DNA microarray; Set (abstract data type); Software; Genetics; Reference genome; Sequence assembly; Gene prediction; Computer science","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.005407541,0.002370239,0.00175819,0.006257944,0.001592896,0.002279546,0.002482891,0.001703003,0.003481058],"category_scores_gemma":[0.01043143,0.001171486,0.001709836,0.002794155,0.0009483967,0.001692185,0.002225604,0.0017404,0.001917688],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008374139,"about_ca_system_score_gemma":0.00116712,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001055686,"about_ca_topic_score_gemma":0.002572912,"domain_scores_codex":[0.9959643,0.0008243853,0.0005394454,0.001156639,0.001298923,0.0002163],"domain_scores_gemma":[0.9952376,0.001788763,0.001145894,0.0006564878,0.0009942085,0.0001770768],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001101756,0.0001337877,0.004902553,0.002271945,0.0005307206,0.0007256152,0.001005591,0.003435387,0.8958577,0.003276316,0.004284769,0.08247384],"study_design_scores_gemma":[0.00017522,0.00111804,0.02493425,0.0004218148,0.0008623699,0.003540501,0.0004908976,0.07992318,0.8312895,0.002912537,0.05384592,0.0004857928],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.125153,0.003056122,0.8330355,0.0001447759,0.000516421,0.0005544142,0.005085953,0.02934112,0.003112633],"genre_scores_gemma":[0.09442125,0.0008230961,0.8923389,0.000122671,0.00003762722,0.0008706978,0.006521543,0.003454656,0.00140959],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.006257944,"threshold_uncertainty_score":0.02859813,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06177281625086602,"score_gpt":0.3084977016402191,"score_spread":0.2467248853893531,"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."}}