We are interested in developing and utilizing new experimental and computational methods in order to understand microbial communities.  Broadly, we are interested in the following areas of research:

  • developing and using bioinformatics tools that better characterize microbial diversity from high-throughput sequencing data
  • developing and using bioinformatics tools for genome assembly and binning of very large metagenomic datasets
  • understanding the genomic determinants of strain-level niche partitioning within microbial communities
  • understanding the genomic determinants of ecosystem-scale processes within microbial communities

We work with a variety of collaborators on specific experimental systems of interest.  Some of these projects are listed below.

EMIRGE: specialized assembly of 16S rDNA and other genes from microbial community sequencing data

We continue to develop EMIRGE (expectation maximization iterative reconstruction of genes from the environment) as a tool for assembly of 16S genes from short-read sequencing data (see Software page).  Even with modern read lengths and error rates of Illumina sequencing, assembling 16S genes from shotgun metagenomics or amplicon sequencing data is challenging.  The goal of EMIRGE is to assembly near-full-length 16S genes from such data, and estimate gene abundances with high accuracy.

Determinants of microbial contributions to methane emissions in freshwater wetlands

Of the three most potent greenhouse gases, methane emissions are the most difficult to predict. Despite being overall carbon sinks, wetlands represent the largest single source of atmospheric methane.  Yet, relatively little is known about the microbial taxa and pathways responsible for methane flux in freshwater wetlands.  Together with the Wrighton lab at Colorado State University and many other great collaborators, we are generating a high-resolution census of freshwater wetland microbial communities, reconstructing the major metabolisms of microbial communities in freshwater wetlands sediments, and aim to understand and ultimately predict how these communities and metabolisms respond to changing climate, hydrology, and variations in available carbon.  This project is producing a large amount of marker gene, metagenomic, and transcriptomic sequencing across all domains of life.  We are working on methods to simultaneously integrate disparate data types from Bacteria, Fungi, Viruses, and especially the understudied Archaea, in order to understand and predict ecosystem-scale processes in freshwater wetlands.

CollaboratorsKelly Wrighton, The Ohio State University

The Golden Retriever Microbiome

The Golden Retriever Lifetime Study is following  more than 3,000 golden retrievers throughout their lifetimes, mining a large amount of health, environmental and behavior data for associations with canine disease, especially cancer.  We are partnering with the Morris Animal Foundation to collect data on the canine fecal microbiome, and analyzing this data in the context of the larger study.

Collaborators: Missy Simpson, Erin Searfoss, Sharon Albright, Nancy Clark, and Barb Wolfe; Morris Animal Foundation

The role of the microbiome in modulating effects of environmental pollutant exposure

Many compounds are discharged from wastewater and agricultural effluents into the environment, with the potential to alter the gut microbiome in fish, other wildlife, companion animals, and humans.  The nature of these alterations and the potential effects on host organisms are unknown.  Using 16S sequencing and metagenomics, we are attempting to characterize pollutant-mediated disruption in the fish gut microbiome, and to estimate the effects of disruption and microbially-mediated contaminant transformation on fish physiology and health.

Collaborators: Alan Vajda, CU Denver; Timberley Roane, CU Denver