2026-10-12T00:08:42Z https://api.figshare.com/v2/oai

oai:figshare.com:article/33977036 2026-09-23T17:51:29Z category_13 category_21 category_272 category_734 category_931 category_811 portal_5 item_type_3 month_year_09_2026

<p>Clustering details from recent TB studies.</p> Yu Lan (132241) Chieh-Yin Wu (22678342) Hsien-Ho Lin (67880) Ted Cohen (118302) Joshua L. Warren (3797263) Genetics Biotechnology Environmental Sciences not elsewhere classified Biological Sciences not elsewhere classified Information Systems not elsewhere classified Mathematical Sciences not elsewhere classified somewhat narrower distribution mycobacterium tuberculosis </ high dyad count div >< p credible interval coverage comparing subsampling strategies pairwise bayesian modeling conduct simulation studies compare two approaches efficient pairwise analysis large pathogen genomic simulation studies control approaches conduct given pairwise analyses pairwise analysis single city shared membership recommend using performed comparably large dataset isolates collected fewer outliers expected 95 entire dataset effect sizes control approach conquer approach computationally intensive computational challenges bias estimates approach mitigates <div><p>Pairwise analysis of genomic and spatial data offers opportunities to identify and estimate the associations between covariates and the transmission of pathogens between individuals. However, such pairwise analyses are computationally intensive, and may not be feasible to conduct given the high dyad count in even moderately sized datasets. Here we compare two approaches to increase the efficiency of pairwise analysis for large datasets. We quantify and compare the performance of divide-and-conquer Bayesian model fitting and pairwise case-control approaches for estimating associations between individual- and pair-level covariates and shared membership in a transmission cluster. We utilize a large dataset (n = 4,154) of spatially-referenced, genomically-sequenced <i>Mycobacterium tuberculosis</i> isolates collected from a single city for this analysis, and conduct simulation studies under three representative tuberculosis genomic clustering settings. Across all simulation studies, the case-control approach produced negligible bias and expected 95% credible interval coverage, and performed comparably to the divide-and-conquer approach, with a somewhat narrower distribution of bias estimates and fewer outliers when effect sizes were large. Thus, we recommend using the case-control approach with five controls per case to downscale datasets for pairwise analysis when analysis of the entire dataset is not possible. This approach mitigates the computational challenges of pairwise Bayesian modeling on datasets that require significant computational resources while maintaining desired inferential properties.</p></div> 2026-09-23T17:53:40Z Dataset Dataset 10.1371/journal.pcbi.1014353.s001 https://figshare.com/articles/dataset/_p_Clustering_details_from_recent_TB_studies_p_/33977036 CC BY 4.0