Multivariate Testing and Modeling for Longitudinal Effects of Maternal HIV-Exposure on Nasopharyngeal Infant Microbiomes in Zambia

• Activity Number: 156 - Contributed Poster Presentations: Section on Statistics in Genomics and Genetics
• Type: Contributed
• Date/Time: Monday, August 8, 2022 : 10:30 AM to 12:20 PM
• Sponsor: Section on Statistics in Genomics and Genetics
• Author(s): Aubrey R. Odom-Mabey*
• Companies: Boston University
• Keywords: Microbiome; Multivariate statistics; HIV; Generalized estimating equations; Mixed modeling; Genomics

Abstract

Worldwide, over 1 million HIV-exposed uninfected (HEU) infants are born to HIV-infected women every year. We collected and analyzed nasopharyngeal swabs from HEU (n=10) and healthy HIV-unexposed uninfected (HUU; n=10) infants and their mothers from Zambia, processed with 16S ribosomal DNA amplicon sequencing, to compare differences in their microbiota over time. Infants were profiled 7 times at 2-3 week intervals over 3.5 months to track post-birth patterns, with two maternal swabs also collected. Multivariate paired Hotelling’s T-squared tests were used to compare microbiota profiles in paired infants at each timepoint, and to compare profiles of HIV+/HIV- mothers-infant pairs. Coefficient estimates for time and HIV exposure were obtained using generalized estimating equations adjusted for clustering of subject samples. We found that some subtle differences do exist between the microbiota of the sampled groups, and also that the nasopharyngeal microbial community changes rapidly over the surveyed 3.5 month period. The research may serve as an insight into existing influences of HIV exposure on infant pulmonary health in early childhood.