Abstract

Abstract: We develop a statistical model to analyse microbiome profiling data based on sequencing of genetic fingerprints in 16S ribosomal RNA . The analysis allows us to quantify the uncertainty in ecological ordination and clustering methods commonly applied in microbiome research. The method is based on the estimation of the underlying microbial distribution in experimental samples using a dependent Dirichet Process prior in which dependence is expressed through low-dimensional latent features. This type of model is advantageous for several reasons. First, information is borrowed across samples to estimate underlying microbial distributions. Second, the nonparametric nature of the model avoids the artefacts of truncation and rarefaction techniques. Lastly, the Bayesian framework mitigates the effects of multiple testing for differential abundance and other hypotheses of interest.

Bio: Sergio Bacallado is a lecturer in the Statistical Laboratory, in the Dept. of Pure Mathematics and Mathematical Statistics at the University of Cambridge. He completed a PhD in Structural Biology at the Stanford Medical School, followed by a research fellowship in the Department of Statistics at Stanford. His research interests include the analysis of dynamical simulations, such as Molecular Dynamics, and applications of Bayesian methods to infer biological mechanisms from high-dimensional, heterogeneous data.