US-UK Collab: Mycobacterial Transmission Dynamics in Agricultural Systems: Integrating Phylogenetics, Epidemiology, Ecology, and Economics

Mathematical modeling of disease transmission is an important tool in studying infectious disease control; however, parameter estimation from field data is often difficult to quantify, especially with complex diseases. Whole-genome sequencing has become faster and more affordable in recent years, thanks to Next-Generation Sequencing techniques. As large databases of sequenced pathogen isolates become available, infectious disease epidemiologists will be able to use these databases, which are more fine-grained than previous serotyping methods, to improve our understanding of transmission dynamics. However, a general methodology is not yet available for such detailed data in bacteria with slow transmission timescales. This project proposes to develop a quantitative methodology for incorporating whole genome sequence (WGS) data into bacterial transmission models. For this, isolates will be used from mycobacterial infections in agricultural systems in the US and UK. Mycobacterial infections, including Mycobacterium avium subsp. paratuberculosis (MAP) and M. bovis (bovine tuberculosis or bTB), are usually studied in simple one-host systems. Mycobacterial disease is extremely difficult to control due to long latent periods, poor diagnostic sensitivity, wildlife and environmental reservoirs of infection, and heterogeneous strain infectiousness. The key to controlling these diseases is believed to be an integrated approach to understand the pathways through which pathogen transmission occurs at all levels in an ecosystem: within animals, between individual animals, between livestock and wildlife, and between livestock and the environment. A generalized, multi-scale methodology is proposed for determining transmission dynamics using WGS data and phylodynamics. These dynamics will be incorporated into predictive models of disease transmission and control within an economic framework, to help decision-makers make informed control choices.

We will develop a new multi-scale model of infection transmission and control which will be used to better understand the principles, dynamics, and impacts of multi-host infections (particularly mycobacterial infections) at multiple levels. These models will be developed in an integrated, phylodynamic parameter estimation framework, using mathematical models that incorporate both our detailed farm and phylogenetic data. In particular, our work will provide a general framework for modeling infection dynamics of slow-moving pathogens within livestock herds and the connection to both wildlife populations and environmental reservoirs that can affect the dynamics of transmission and control. Our work will be based on unique and innovative data collection methods and data analyses. We will test our methods using expansive isolate collections and precise data on two mycobacterial infections: Mycobacterium avium subsp. paratuberculosis (MAP), the cause of Johne's disease (JD) in ruminants, and Mycobacterium bovis (M. bovis), the causal agent of bovine tuberculosis (bTB). Bovine tuberculosis will be studied in cattle and white-tailed (WT) deer (Odocoileus virginianus) in the US and compared to bTB in cattle and Eurasian badgers (Meles meles) in the UK, while MAP will be studied in dairy cattle and their environment in both the US and UK. These infections are active in complex ecosystems and affect human populations through direct contact, economic impact, and potential contamination of the food supply. By comparing across pathogens, we can examine the impact of the difference between wildlife reservoirs (M. bovis) and environmental reservoirs (MAP), with their differing mutation rates4 while determining the general principles involved in integrating WGS and epidemiological models.

Transformative research in phylodynamics and disease ecology is proposed to improve our understanding of bacterial infection dynamics by developing 1) a generalized methodology for incorporating WGS data into transmission models, 2) a generalized ecosystem approach to modeling infectious diseases with agricultural production and economic components, and 3) research methods for modeling transmission dynamics in complicated multi-scale systems. This project will integrate innovative phylodynamic techniques with novel data collection and analysis across various scales in the US and UK to determine the minimum necessary data requirements for this model type. The results will allow for design of cost-effective data collection programs, which will be valuable in develop countries but essential in many livestock-dependent low-income countries.

phylodynamic framework at the core of this work will have broad use in public health policy and practices through identifying important factors in multi-host, systems; the generality of the framework will allow it to be adapted broadly to bacterial diseases affecting humans and animals. This interdisciplinary work will promote a cross-fertilization of ideas amongst students, researchers and faculty in molecular biology, epidemiology, ecology, and economics. The molecular biologists, epidemiologists, ecologists and economists in the team will collaborate closely to develop the techniques for this work. Other research-related education and outreach activities specific to this work will include workshops at NIMBioS, short-courses on transmission for veterinarians, and an intensive graduate-level course on endemic disease modeling. Furthermore, members of the group actively participate in science educational programs for 7th-9th grade girls in the Ithaca, NY community and in education for school-age children in the UK, as well as professional development classes in many countries.

At the completion of the project, a methodology will have been developed for multi-scale generalized systems modeling of multi-host infectious diseases in an ecosystem that includes livestock, environmental, wildlife, and economic components with a particular application in mycobacterial diseases and their control. A large impact is expected in core phylogenetic and mathematical epidemiology areas, including multi-scale modeling and inference from WGS data. In the process high school, undergraduates, graduate students and post-doctoral fellows will have been trained and enthused in the use of this approach.