Differential abundance analysis for microbial marker-gene surveys

1] Graduate Program in Applied Mathematics & Statistics, and Scientific Computation, University of Maryland, College Park, Maryland, USA. [2] Center for Bioinformatics and Computational Biology, University of Maryland, College Park, Maryland, USA.
Nature Methods (Impact Factor: 25.95). 09/2013; 10. DOI: 10.1038/nmeth.2658
Source: PubMed

ABSTRACT We introduce a methodology to assess differential abundance in sparse high-throughput microbial marker-gene survey data. Our approach, implemented in the metagenomeSeq Bioconductor package, relies on a novel normalization technique and a statistical model that accounts for undersampling-a common feature of large-scale marker-gene studies. Using simulated data and several published microbiota data sets, we show that metagenomeSeq outperforms the tools currently used in this field.

Download full-text


Available from: Joseph Paulson, Feb 17, 2015
1 Follower
  • [Show abstract] [Hide abstract]
    ABSTRACT: Nucleotide sequencing has become increasingly common and affordable, and is now a vital tool for studies of the human microbiome. Comprehensive microbial community surveys such as MetaHit and the Human Microbiome Project have described the composition and molecular functional profile of the healthy (normal) intestinal microbiome. This knowledge will increase our ability to analyze host and microbial DNA (genome) and RNA (transcriptome) sequences. Bioinformatic and statistical tools can then be used to identify dysbioses that might cause disease, and potential treatments. Analyses that identify perturbations in specific molecules can leverage thousands of culture-based isolate genomes to contextualize culture-independent sequences, or may integrate sequence data with whole-community functional assays such as metaproteomic or metabolomic analyses. We review the state of available systems-level models for studies of the intestinal microbiome, along with analytic techniques and tools that can be used to determine its functional capabilities in healthy and unhealthy individuals.
    Gastroenterology 01/2014; 146. DOI:10.1053/j.gastro.2014.01.049 · 13.93 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The composition of the human gut microbiome is influenced by many environmental factors. Diet is thought to be one of the most important determinants, though we have limited understanding of the extent to which dietary fluctuations alter variation in the gut microbiome between individuals. In this study, we examined variation in gut microbiome composition between winter and summer over the course of one year in 60 members of a founder population, the Hutterites. Because of their communal lifestyle, Hutterite diets are similar across individuals and remarkably stable throughout the year, with the exception that fresh produce is primarily served during the summer and autumn months. Our data indicate that despite overall gut microbiome stability within individuals over time, there are consistent and significant population-wide shifts in microbiome composition across seasons. We found seasonal differences in both (i) the abundance of particular taxa (false discovery rate <0.05), including highly abundant phyla Bacteroidetes and Firmicutes, and (ii) overall gut microbiome diversity (by Shannon diversity; P = 0.001). It is likely that the dietary fluctuations between seasons with respect to produce availability explain, at least in part, these differences in microbiome composition. For example, high levels of produce containing complex carbohydrates consumed during the summer months might explain increased abundance of Bacteroidetes, which contain complex carbohydrate digesters, and decreased levels of Actinobacteria, which have been negatively correlated to fiber content in food questionnaires. Our observations demonstrate the plastic nature of the human gut microbiome in response to variation in diet.
    PLoS ONE 03/2014; 9(3):e90731. DOI:10.1371/journal.pone.0090731 · 3.53 Impact Factor
  • Nature Methods 03/2014; 11(4):359-60. DOI:10.1038/nmeth.2898 · 25.95 Impact Factor
Show more