{"id":"W4393276601","doi":"10.21203/rs.3.rs-4159693/v1","title":"Learning novel SARS-CoV-2 lineages from wastewater sequencing data","year":2024,"lang":"en","type":"preprint","venue":"Research Square","topic":"SARS-CoV-2 detection and testing","field":"Medicine","cited_by":2,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Waterloo","funders":"University of Waterloo; Mitacs; Ministry of Environment; Queen's University; Trent University; University of Ottawa; Public Health Agency; Public Health Agency of Canada; Ministère de l’Environnement, de la Protection de la nature et des Parcs; Northwestern University","keywords":"Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2); Coronavirus disease 2019 (COVID-19); 2019-20 coronavirus outbreak; Wastewater; Sars virus; Biology; Computational biology; Virology; Computer science; Medicine; Environmental science; Infectious disease (medical specialty); Environmental engineering","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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","research_integrity","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.00295879,0.0004814805,0.0007376282,0.0009527446,0.0003239302,0.0005641123,0.0009165296,0.0006559159,0.000104292],"category_scores_gemma":[0.005102673,0.0004186967,0.0002236209,0.0007130931,0.0002058894,0.000130687,0.007875146,0.008506412,0.001220666],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006965646,"about_ca_system_score_gemma":0.001266462,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005253629,"about_ca_topic_score_gemma":0.0002298249,"domain_scores_codex":[0.994303,0.0003882117,0.0006689792,0.001856569,0.001790213,0.0009929802],"domain_scores_gemma":[0.995684,0.0009533553,0.0001346658,0.00240879,0.0006744991,0.0001446723],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0001054354,0.0000820581,0.0009046558,0.002401536,0.0003173699,0.0007993617,0.001227446,0.00004929391,0.9851449,0.00001750961,0.003147811,0.005802589],"study_design_scores_gemma":[0.0009134488,0.0002990598,0.0002280297,0.009364589,0.000174433,0.0001416162,0.001947938,0.07543834,0.8740623,0.0007967435,0.03610248,0.0005310201],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.986842,0.002389178,0.000265924,0.001174226,0.0008965161,0.001069553,0.000437692,0.0008199582,0.006104902],"genre_scores_gemma":[0.9897532,0.00009946396,0.004107052,0.0003042388,0.002749933,0.00009900348,0.001360676,0.0002235369,0.001302827],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.1110826,"threshold_uncertainty_score":0.9998265,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.3272777793403889,"score_gpt":0.4667492163911954,"score_spread":0.1394714370508066,"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."}}