Math and the Microbiome

From the bacteria in our guts, to microbes involved in biodegradation and crop growth, to viruses in the ocean, some of Earth’s tiniest organisms play some of the most important roles in global health, food production, and climate change. Advances in metagenomic sequencing technology including 16S, viromics, and mycobiomics - along with metabolomics, transcriptomics, and proteomics allow us to characterize these complex microbial communities and begin to understand their functions. This Big Data creates opportunities for data driven discovery and new data analytics, but Big Data also comes with challenges: Meaningful integration of multi-omic data has become increasingly critical to microbiome studies as recent work highlights the importance of community dynamics, interactions, and microbial ecology over the roles of individual microbes. For example, microbial metabolisms are now recognized to often be ‘distributed’ across consortia; viruses manipulate microbial metabolisms and population dynamics, and co-occurring fungi in most ecosystems are virtually unstudied but likely play key roles as well. Data integration techniques range from correlations to network analyses to genome-scale microbial community metabolic models that assess metabolite flux to ecosystem models that provide predictive power of which organisms drive key features of the system. Some of these techniques, like correlations, accommodate many types of –omic data but cannot account for the complex biology or ecology of a system. Other techniques, like metabolic modeling, better account for this complexity, but do not yet integrate phenotypic –omic data (i.e. metabolomics, proteomics) well. Each of these techniques has advantages and limitations and new computational tools for data integration and modeling have rapidly developed over the last 2 years. Besides data integration, Whether studying environmental, gut, or industrial microbes, the ability to accurately identify and predict the structure and function of microbial communities has far-reaching potential and paves the way for microbial engineering in bioremediation, probiotic development, and sustainable agriculture.

In this 3 day workshop, we will take a genome to phonome approach with broad perspectives provided by mathematicians, biologists, and statisticians. We will also develop interdisciplinary working subgroups to consider the questions, challenges, tools, and needs of data integration and modeling in microbiome studies. Each participant will present a short talk (5 minutes, 3 slides) highlighting his or her research, perspectives, and challenges. The goal is to help develop a broadly collaborative community of math-enabled microbiome scientists with common research goals.

This workshop is co-sponsored by the Mathematical Biosciences Institute, the National Institute of Statistical Sciences, and the Infectious Diseases Institute at Ohio State University.