{"id":"W2950712771","doi":"10.1101/060186","title":"Using reference-free compressed data structures to analyse sequencing reads from thousands of human genomes","year":2016,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Genomics and Phylogenetic Studies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto; Ontario Institute for Cancer Research","funders":"Wellcome Trust","keywords":"1000 Genomes Project; Population; Reference genome; Genome; Human genome; Indel; Computer science; Computational biology; SNP genotyping; Sequence assembly; Biology; Data mining; Genetics; Genotyping; Single-nucleotide polymorphism; Genotype; 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":[0.001313687,0.0004678652,0.0004214124,0.001414454,0.0004245071,0.0009718238,0.0008382186,0.0006118002,0.001730202],"category_scores_gemma":[0.006727315,0.0002901619,0.0004466803,0.002671597,0.0003593549,0.0009935749,0.0007399636,0.000604569,0.000955243],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005530669,"about_ca_system_score_gemma":0.0006013975,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002092506,"about_ca_topic_score_gemma":0.002034444,"domain_scores_codex":[0.9991125,0.0001891693,0.00009932393,0.0002236931,0.000330302,0.00004493913],"domain_scores_gemma":[0.997732,0.0009840891,0.000228506,0.0004911066,0.0005068989,0.0000573392],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.003626763,0.0003938513,0.02409038,0.001054538,0.000446409,0.001257673,0.001274935,0.0797536,0.4003497,0.01359208,0.01852348,0.4556366],"study_design_scores_gemma":[0.0003988497,0.0008145045,0.02859205,0.0001602381,0.000172254,0.001360762,0.0006822862,0.5826497,0.3266691,0.02580593,0.0325322,0.0001621402],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5863525,0.001473784,0.3766971,0.000883791,0.0002321279,0.0001832743,0.01719373,0.01413225,0.002851533],"genre_scores_gemma":[0.5389279,0.0004599126,0.4267835,0.0002340518,0.00008095798,0.0002850259,0.03087437,0.0008059503,0.001548346],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.002092506,"threshold_uncertainty_score":0.006947517,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07230040717983077,"score_gpt":0.2895389052507873,"score_spread":0.2172384980709566,"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."}}