Multi-sample estimation of centered log-ratio matrix in microbiome studies

TL;DR

This paper introduces a multi-sample estimation method for the centered log-ratio matrix in microbiome studies, addressing zero counts and improving compositional analysis accuracy.

Contribution

It proposes a regularized maximum likelihood estimator with nuclear norm penalty for clr matrices, leveraging low-rank structure across multiple samples.

Findings

Outperforms naive estimators in simulations

Effective on real microbiome datasets

Provides theoretical error bounds

Abstract

In microbiome studies, one of the ways of studying bacterial abundances is to estimate bacterial composition based on the sequencing read counts. Various transformations are then applied to such compositional data for downstream statistical analysis, among which the centered log-ratio (clr) transformation is most commonly used. Due to limited sequencing depth and DNA dropouts, many rare bacterial taxa might not be captured in the final sequencing reads, which results in many zero counts. Naive composition estimation using count normalization leads to many zero proportions, which makes clr transformation infeasible. This paper proposes a multi-sample approach to estimation of the clr matrix directly in order to borrow information across samples and across species. Empirical results from real datasets suggest that the clr 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 is developed. Theoretical upper bounds of the estimation errors and of its corresponding singular subspace errors are established. Simulation studies demonstrate that the proposed estimator outperforms the naive estimators. The method is analyzed on Gut Microbiome dataset and the American Gut project.