Research 

My research group uses “big data” and “multi-omic” approaches to understand and engineer the microbiomes of built and natural environments to preserve both human and environmental health.

We currently have active projects in three main areas:

1. developing novel phage-based drinking water treatment technologies 

2. characterizing what makes a healthy indoor microbiome 

3. reducing greenhouse gas emissions and pollutant runoff from wetlands 

If you are interested in collaborations (whether in these or other related areas), email me. I’m actively recruiting new undergraduate students, graduate students, and postdocs. 

Characterizing Drinking Water Phages

Bacteriophages, better known as phages, are viruses that infect bacteria. In natural environments, they are known to be critical for regulating microbial population dynamics; however, the interactions between bacteria and viruses in drinking water are largely unknown. Previous studies have employed metagenomics and amplicon sequencing to understand bacteria that grow in drinking water at various stages of treatment, within distribution systems, and in premise plumbing, leaving unexplored the importance of viruses at each of these stages.

By mining publicly available drinking water metagenomes, I was able to assess the dynamics of viral populations in drinking water distribution systems from around the world for the first time. I revealed that drinking water phages vary in abundance based on the characteristics of the drinking water distribution system, particularly based on residual disinfectant use. Drinking water distribution systems that use a residual disinfectant are less diverse and less even than those distribution systems that do not use a residual disinfectant. Additionally, drinking water viruses carry genes to survive the conditions of oxidative stress and nitrogen limitation in drinking water distribution systems.

Improving Viral sequence Identification Pipelines

Numerous tools exist for filtering viral sequences from mixed metagenomes. These tools rely on protein similarity and/or machine learning approaches. However, it remains difficult to accurately identify viral sequences from mixed metagenomes. One possible way to improve viral sequence identification accuracy may be to combine the output of multiple tools to create a more high confidence set of viral sequences.

I am currently leading a team to assess the effect of combining the output from multiple tools (VirSorter, VirSorter2, CheckV, DeepVirFinder, VirFinder, VIBRANT, and Kaiju) on viral sequence prediction. Preliminary data suggests improved precision and recall by combining at least 4 of the tools. Upon finalizing our prediction pipeline, all scripts will be made available on github.

MycobacteriA-Phage Interactions in Drinking Water

The recent rise in nontuberculosis mycobacteria (NTM) infections in many cities has demonstrated the need to develop novel approaches for controlling NTM presence in tap water. Harnessing mycobacteriophages (viruses that infect mycobacteria but are harmless to humans) may be one possible way to do so. However, the presences of mycobacteriophages in drinking water remains unknown.

To fill this gap, it is essential to develop better ways of concentrating viruses from drinking water to be able to connect the abundances of NTMs and their phages in drinking water distribution systems. My postdoctoral work seeks to correlate the presence of NTMs and mycobacteriophages in the Ann Arbor distribution system. This research will provide essential information about the mycobacteriophages present in drinking water and their target host range and is critical to assess the feasibility of designing a mycobacteriophage-based method for controlling NTM abundance in drinking water systems.

Fungal Community Gene Expression

Dampness and visible mold are ubiquitous in buildings. Nearly 1/5 of buildings in North America and Europe have problems with dampness. The problem is even worse in homes, where an estimated half of US homes are considered damp. This dampness and visible mold are known to lead to negative health effects; however, the casual relationships between fungal exposure and health impacts remains nebulous.

I am investigating these connections; seeking to understand the ecology and activity of the microbial communities that also inhabit our homes. In particular, I have demonstrated that dampness impacts fungal community gene expression, something that previous studies have overlooked. In future work, I will seek to show how what the community is doing is equally, if not more important, than what the community’s composition is.

Classifying Moldy Homes

Visual inspection remains one of the most relied-upon methods for assessing whether a home has a mold problem or not. Reflecting the fact that mold often is found in hidden spaces in  a home, PCR-based methods, as well as culturing and microscopy, are also used. However, these do not adequately represent the diversity of fungi found in a building and have to be modified to account for how community profile and total fungal abundances vary between different geographic regions and with changing seasons. For these reasons, accurately identifying whether or not a home has mold damage remains challenging.

My research is laying the groundwork for the development of a quantitative and accurate tool that mold inspectors and homeowners alike can use to diagnose homes with mold damage. Using a high throughput DNA sequencing and computational biology approach, I have identified ecological differences between the fungal communities of the settled dust of homes with reported moisture or visible mold versus homes with no known mold growth or moisture problems (normal ecology). Recognizing the profound influence of outdoor fungal ecology on that of the indoors, these samples were taken from homes in a variety of US climate regions. An intensive single-home study compliments the national effort and reveals that mold from damp material can be spread throughout a single-family home.

Cyanobacteria-derived Biofuels

As the effects of climate change become more pronounced, it is clear that we need to break our addiction to fossil fuels. Cyanobacteria-derived biofuels are one promising alternative to our need for a sustainable drop-in fuel. As biofuels are capable of being burned by the vehicles already on the road, they would not require a massive turnover in the transportation sector, reducing a significant barrier to other green technologies. Furthermore, cyanobacteria are capable of producing other chemicals, such as pigments and pharmaceuticals, that can help offset their cost.

Through my research, I am developing a cyanobacteria platform for the production of biofuel-precursors and valuable chemical products, using the strain Synechococcus elongatus UTEX 2973.

To read more about my past work, check out my google scholar.