{"id":"W4287549685","doi":"10.5281/zenodo.4386962","title":"Demonstrating the utility of flexible sequence queries against indexed short reads with FlexTyper","year":2020,"lang":"en","type":"article","venue":"Zenodo (CERN European Organization for Nuclear Research)","topic":"Algorithms and Data Compression","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"","keywords":"Sequence (biology); Computer science; Information retrieval; Biology; Genetics","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":[0.0004048265,0.0001042091,0.0001160878,0.0000406474,0.001126931,0.0004448473,0.001605163,0.00003330762,0.0006806541],"category_scores_gemma":[0.0001913592,0.00007373835,0.00002582933,0.0006196341,0.0002348737,0.0005413272,0.001478728,0.0002177355,0.0002097716],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001900058,"about_ca_system_score_gemma":0.00001142936,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001566013,"about_ca_topic_score_gemma":1.874265e-7,"domain_scores_codex":[0.9986505,0.0002005482,0.000201367,0.0003557561,0.0003746817,0.0002171496],"domain_scores_gemma":[0.9988672,0.00002720191,0.00007918295,0.0005636421,0.0003427148,0.0001200664],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0001697891,0.0001637442,0.0001708497,0.0001162869,0.00008100127,0.0000472028,0.008705875,0.0003337627,0.02277835,0.03936883,0.0686845,0.8593798],"study_design_scores_gemma":[0.0005541214,0.0008255013,0.005291048,0.0001265113,0.00001794381,0.0001327973,0.00152853,0.1727546,0.01350696,0.0003260868,0.8044541,0.0004817716],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.04387663,0.0001265458,0.8427548,0.003820342,0.0000914121,0.000640134,0.0001518775,0.001320028,0.1072183],"genre_scores_gemma":[0.9929186,0.00001689627,0.006341372,0.000318779,0.000046891,4.134648e-8,0.0001343153,0.0001704276,0.00005268592],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.949042,"threshold_uncertainty_score":0.8667556,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0796312605441998,"score_gpt":0.2603976221947847,"score_spread":0.1807663616505849,"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."}}