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: 32.07). 09/2013; 10(12). DOI: 10.1038/nmeth.2658
Source: PubMed


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
  • Source
    • "Alpha diversity (Shannon Index) was calculated on unfiltered data. Read counts were normalized using metagenomeSeq (Paulson et al., 2013) which utilizes a cumulative-sum scaling where raw counts are divided by the cumulative sum of counts up to a particular quantile. Principal Coordinate Analysis (PCoA) using Bray-Curtis dissimilarity indices, analysis of differential abundance of taxa, and beta diversity analysis using Whittaker's species turnover were performed on filtered and normalized data. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Rodent models of arthritis have been extensively used in elucidation of rheumatoid arthritis pathogenesis and they are instrumental in development of therapeutic strategies. Here we utilise delayed-type hypersensitivity arthritis (DTHA), a model in C57BL/6 mice affecting one paw with synchronised onset, 100% penetrance and low variation. We investigate the role of Tregs in DTHA through selective depletion of Tregs and the role of IL-17 in connection with Treg depletion. Given the relevance of Tregs in RA, and the possibility of developing Treg directed therapies, this approach could be relevant for advancing the understanding of Tregs in inflammatory arthritis. Selective depletion of Tregs was achieved using a FoxP3-DTR-eGFP mouse that expresses the diphtheria toxin receptor and eGFP under control of the foxp3 gene. Anti-IL-17 mAb was used for IL-17 blockade. Numbers and activation of Tregs increased in the paw and its draining lymph node in DTHA and depletion of Tregs resulted in exacerbation of disease shown by increased paw swelling, increased infiltration of inflammatory cells, increased bone remodelling and increased production of inflammatory mediators as well as increased production of anti-citrullinated protein antibodies. Anti-IL-17 mAb treatment demonstrated that IL-17 is important for disease severity in both presence and absence of Tregs and that IL-17 blockade is able to rescue mice from the exacerbated disease caused by Treg depletion and caused reduction in RANKL, IL-6 and neutrophil numbers. We show that Tregs are important for the containment of inflammation and bone remodelling in DTHA and that IL-17 is an important driver of disease both in presence and absence of Tregs. To our knowledge this is the first study using the FoxP3-DTR-eGFP mouse on a C57BL/6 background for Treg depletion in an arthritis model, and we here demonstrate the usefulness of the approach to study the role of Tregs and IL-17 in arthritis.
    Preview · Article · Jan 2016 · Disease Models and Mechanisms
  • Source
    • "Differential abundance analyses were conducted on the normalized OTU table using the 'fitZig' test in MetagenomeSeq on all OTUs with a variance threshold greater than 30 (Paulson et al., 2013). Results focus on classified clades (unless at the OTU level) with a relative abundance greater than 1% in at least one phase and all p-values were adjusted for multiple comparisons using the false discovery rate (Benjamini and Hochberg, 1995). "
    [Show abstract] [Hide abstract]
    ABSTRACT: Unprecedented insect-induced tree mortality has been observed globally and while hydrologic and biogeochemical changes have been recorded, alterations to terrestrial microbial communities, which influence as well as respond to these shifts, are not well understood. The objective of this work was to better understand how bacterial communities are coupled to perturbations in biogeochemically-relevant soil physicochemical parameters resulting from beetle-induced tree death. To this end, soils beneath trees across a beetle-kill spectrum were contrasted in the central Rocky Mountains at both heavily impacted (Chimney, 85% mortality) and moderately impacted (Niwot, 18% mortality) field sites. Soil organic matter (OM), dissolved organic carbon (DOC) and dissolved organic nitrogen were all significantly altered when contrasting soils under healthy versus beetle-attacked trees at Chimney. Bacterial alpha diversity measurements were found to increase with tree death and beta diversity measures showed significant clustering with relation to tree phase. The site was characterized by a significantly higher relative abundance of bacterial clades under healthy trees that were correlated to OM and DOC concentrations. In contrast, compositional changes in soil bacterial communities and edaphic parameters associated with tree phase were not observed at the less impacted Niwot site. Our findings reveal a coupled response between shifts in organic carbon cycling and the bacterial assemblage as a result of large-scale, beetle-induced tree mortality with implications for heterotrophic respiration in near-surface soils and suggests a possible dependence on the level of forest mortality before manifestation.
    Full-text · Article · Dec 2015 · Soil Biology and Biochemistry
    • "Sequences representing any phylotypes classified as mitochondria or chloroplast were removed. In order to reduce potential amplicon sequencing biases, we first removed samples with < 10 000 sequences and then we normalised the sequence counts, using a cumulative-sum scaling approach (Paulson et al. 2013). The total number of samples included in downstream analyses was 556 for bacteria and 480 for fungi. "
    [Show abstract] [Hide abstract]
    ABSTRACT: The complexities of the relationships between plant and soil microbial communities remain unresolved. We determined the associations between plant aboveground and belowground (root) distributions and the communities of soil fungi and bacteria found across a diverse tropical forest plot. Soil microbial community composition was correlated with the taxonomic and phylogenetic structure of the aboveground plant assemblages even after controlling for differences in soil characteristics, but these relationships were stronger for fungi than for bacteria. In contrast to expectations, the species composition of roots in our soil core samples was a poor predictor of microbial community composition perhaps due to the patchy, ephemeral, and highly overlapping nature of fine root distributions. Our ability to predict soil microbial composition was not improved by incorporating information on plant functional traits suggesting that the most commonly measured plant traits are not particularly useful for predicting the plot-level variability in belowground microbial communities.
    No preview · Article · Oct 2015 · Ecology Letters
Show more