{"id":"W4214559301","doi":"10.1145/3503222.3507702","title":"GenStore: a high-performance in-storage processing system for genome sequence analysis","year":2022,"lang":"en","type":"article","venue":"","topic":"Genomics and Phylogenetic Studies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":68,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Computer science; Heuristics; Genome; Genomics; Computation; Set (abstract data type); Reference genome; Computational genomics; String searching algorithm; String (physics); Data mining; Computational biology; Pattern matching; Algorithm; Artificial intelligence; Biology; Genetics; Mathematics","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0002125377,0.0001118448,0.000176977,0.00009230831,0.0002261003,0.00001331368,0.0001831669,0.0000332598,0.000008876651],"category_scores_gemma":[0.000004947228,0.0001107101,0.000075383,0.0002984928,0.00002612124,6.499474e-7,0.0001642934,0.00004617497,8.326801e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007267087,"about_ca_system_score_gemma":0.00006363019,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003610456,"about_ca_topic_score_gemma":0.00004122572,"domain_scores_codex":[0.9991469,0.00002475996,0.0001892631,0.0003269032,0.00009116691,0.0002209562],"domain_scores_gemma":[0.9996357,0.000004957813,0.00007120858,0.000209558,0.00004780284,0.00003079498],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"observational","study_design_scores_codex":[0.00009245791,0.0000445208,0.02231535,0.000121755,0.0002954076,0.000003936293,0.0001822534,0.06883989,0.9063097,0.0001791592,0.00004445534,0.001571072],"study_design_scores_gemma":[0.005773634,0.003270041,0.5059763,0.00003232438,0.001492279,0.0001232018,0.005994556,0.07906646,0.1520128,0.000113294,0.2428836,0.003261524],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9962852,0.001194363,0.001568746,0.00003411855,0.0001171328,0.0002328133,0.00008391506,0.00000576286,0.0004779517],"genre_scores_gemma":[0.9975661,0.00004343946,0.001293689,0.00009241434,0.00007149287,0.0002649134,0.0001033983,0.00001418008,0.0005503873],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.754297,"threshold_uncertainty_score":0.4514628,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0134734046518476,"score_gpt":0.2266856231496225,"score_spread":0.2132122184977749,"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."}}