Project Summary:
Zoonotic diseases result from the spillover of pathogens from animal reservoirs to humans due
to contact with animals or animal products. An estimated 60% of known infectious diseases and
up to 75% of emerging infectious diseases are zoonotic in origin. Despite this, relatively little is
known about the alterations to the human microbiome in persons with a high degree of close
contact with animals on farms. We propose a study to understand the pathogenic and non-
pathogenic effects of a high degree of livestock exposure. We will use shotgun metagenomics
to investigate which bacteria colonize the human gut after zoonotic exposure, which bacteria are
outcompeted by colonizers, and which bacteria adapt. To enable comparisons with respect to
virulence, beneficial metabolic pathways and antibiotic resistance, we will develop novel
statistical methods for bacterial pangenomics. Existing methods for bacterial pangenomics
assume that all genomes are observed without error, which rarely is the case for metagenome-
assembled genomes (MAGs). For example, MAGs may omit genes that are truly present in the
target genome, or MAGs may contain erroneously observed genes. To address limitations of
current methods, we will develop statistical methodology that adjusts for differential quality in
gene-level comparisons of metagenome-assembled genomes. Additionally, we will develop
guidelines for the design of shotgun sequencing experiments based on maximizing power to
test hypotheses about associations with gene presence. Successful completion of these aims
will increase our understanding of the mechanisms of transmission of zoonotic pathogens and
commensals, and advance broadly applicable and essential methodology for comparing
bacterial genomes.