Abstract

The human microbiome comprises the microorganisms that inhabit the various locales of the human body and plays a vital role in human health. The composition of a microbial population is often quantified through measures of species diversity, which summarize the number of species along with their relative abundances into a single value. In a microbiome sample there will certainly be species missing from the target population which will affect the diversity estimates. We employ a model based on the hierarchical Pitman-Yor (HPY) process (Battiston et al. 2018) to model the species abundance distributions over multiple microbiome populations. The model parameters are estimated using a Gibbs sampler. We also show how estimates of true population diversity can be obtained as a function of the HPY parameters, and that the estimates from the HPY model improve over naive estimates. Finally, we run the HPY procedure on an infant gut microbiome dataset from the Canadian Healthy Infant Longitudinal Development (CHILD) cohort study.