Abstract
AbstractIt was recently argued1that an analysis of tumor-associated microbiome data2is invalid because features that were originally very sparse (genera with mostly zero read counts) became associated with the phenotype following batch correction1. Here, we examine whether such an observation should necessarily indicate issues with processing or machine learning pipelines. We focus on the centered log ratio (CLR) transformation, which is often recommended for analysis of compositional microbiome data3. The CLR transformation has similarities to Voom-SNM4,5, the batch-correction method brought into question1,2, yet is a sample-wise operation that cannot, in itself, “leak” information or invalidate downstream analyses. We show that because the CLR transformation divides each value by the geometric mean of its sample, common imputation strategies for missing or zero values result in transformed features that are associated with the geometric mean. Through analyses of both synthetic and vaginal microbiome datasets we demonstrate that when the geometric mean is associated with a phenotype, sparse and CLR-transformed features will also become associated with it. We re-analyze features highlighted by Gihawi et al.1and demonstrate that the phenomena of sparse features becoming phenotype-associated can also be observed after a CLR transformation. While we do not intend to validate tumor-associated microbiome signatures2or evaluate other concerns regarding their detection and analysis1,6, we conclude that as phenotype-associated features that were initially sparse can be created by a sample-wise transformation that cannot artifactually inflate machine learning performance, their detection is not independently sufficient to demonstrate an analytic issue in machine learning pipelines. However, as was also previously noted by others, features transformed with sample-wise operations such as the CLR transformation should be interpreted with caution.
Publisher
Cold Spring Harbor Laboratory