Statistical Methods for Detecting Differentially Abundant Features in Clinical Metagenomic Samples

We present a statistical method for comparing clinical metagenomic samples from two treatment populations on the basis of count data to detect differentially abundant features

James Robert White; Niranjan Nagarajan; Mihai Pop


Scholarcy highlights

  • The increasing availability of high-throughput, inexpensive sequencing technologies has led to the birth of a new scientific field, metagenomics, encompassing large-scale analyses of microbial communities
  • Our approach relies on the following assumptions: we are given data corresponding to two treatment populations each consisting of multiple individuals; for each sample we are provided with count data representing the relative abundance of specific features within each sample, e.g. number of 16S rRNA clones assigned to a specific taxon, or number of shotgun reads mapped to a specific biological pathway or subsystem
  • The input to our method can be represented as a Feature Abundance Matrix whose rows correspond to specific features, and whose columns correspond to individual metagenomic samples
  • In order to validate our method, we first designed simulations and compared the results of Metastats to Student’s ttest and two methods used for SAGE data: a log-linear model by Lu et al, and a negative binomial model developed by Robinson and Smyth
  • We designed a metagenomic simulation study in which ten subjects are drawn from two groups - the sampling depth of each subject was determined by random sampling from a uniform distribution between 200 and 1000
  • Lu et al designed a similar study for SAGE data, for each simulation, a fixed dispersion was used for both populations and the dispersion estimates were remarkably small
  • Subsystems for RNA and DNA metabolism were significantly more abundant in viral metagenomes, while nitrogen metabolism, membrane transport, and carbohydrates were all enriched in microbial communities

Need more features? Save interactive summary cards to your Scholarcy Library.