{"id":"W2910562404","doi":"10.1109/bdcat.2018.00014","title":"ntPack: A Software Package for Big Data in Genomics","year":2018,"lang":"en","type":"article","venue":"","topic":"Genomics and Phylogenetic Studies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"BC Cancer Agency","funders":"Canada's Michael Smith Genome Sciences Centre","keywords":"Genomics; Computer science; Software; Computational genomics; Scalability; Shotgun sequencing; Genome; Big data; Sequence assembly; Leverage (statistics); DNA sequencing; Computational biology; Data mining; Hybrid genome assembly; Data science; Biology; Database; Artificial intelligence; Transcriptome; Genetics; Gene","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":true,"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":[0.005131116,0.003209635,0.002679045,0.002763368,0.001406198,0.002926292,0.004955573,0.001280456,0.05684474],"category_scores_gemma":[0.01715099,0.002777394,0.002833654,0.003983914,0.0009917496,0.004244925,0.003791899,0.005340056,0.03797511],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001006845,"about_ca_system_score_gemma":0.003789867,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00508238,"about_ca_topic_score_gemma":0.005363902,"domain_scores_codex":[0.9977705,0.0006477701,0.0003609632,0.0004033514,0.0006324188,0.0001849927],"domain_scores_gemma":[0.9938964,0.003981964,0.0003732505,0.0007955064,0.0006605054,0.0002923891],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.001297427,0.0002204774,0.003156316,0.003811628,0.0008195019,0.0006502604,0.0008651368,0.01229003,0.008809852,0.02077783,0.7729542,0.1743474],"study_design_scores_gemma":[0.001200644,0.0002068706,0.004649363,0.0006502913,0.0003326567,0.0008308248,0.000285588,0.1274931,0.02151104,0.08701126,0.7553786,0.0004498047],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"methods","genre_gemma":"software","genre_scores_codex":[0.001769301,0.0006093124,0.4681545,0.0004606939,0.0004026488,0.0006515391,0.05926083,0.4639637,0.004727547],"genre_scores_gemma":[0.01750412,0.0008671586,0.726424,0.0008139561,0.0002103979,0.004528159,0.1185616,0.1252981,0.005792536],"genre_candidate":"software","genre_consensus":null,"teacher_disagreement_score":0.05684474,"threshold_uncertainty_score":0.1901646,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04912872420318735,"score_gpt":0.2783618497948962,"score_spread":0.2292331255917088,"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."}}