Presentation: Multi-sample Estimation of Bacterial Composition Matrix in Metagenomics Data
Speaker: Hongzhe Li, Ph.D., Professor of Biostatistics and Statistics, University of Pennsylvania Perelman School of Medicine
Abstract: Metagenomics sequencing is routinely applied to quantify the bacterial abundances in microbiome studies, where the bacterial composition is estimated based on the sequencing read counts. Due to limited sequencing depth and DNA dropouts, many rare bacterial taxa might not be captured in the final metagenomic sequencing, which results in many zero counts. Naive composition estimation using count normalization leads to many zero proportions, which tend to result in biased estimates of bacterial abundance and diversity. This paper takes a multi-sample approach to estimation of bacterial abundances in order to borrow information across samples and across species. Empirical results from real data sets suggest that the composition matrix over multiple samples is approximately low rank, which motivates a regularized maximum likelihood estimation with a nuclear norm penalty. An efficient optimization algorithm using the generalized accelerated proximal gradient and Euclidean projection onto the simplex space is developed. The theoretical upper bounds and the minimax lower bounds of the estimation errors, measured by the Kullback-Leibler divergence and the Frobenius norm, are established. Simulation studies demonstrate that the proposed estimator outperforms the naive estimators. The methods is applied to analysis of a human gut microbiome dataset.