Predicting Pathogen Spread Across Mammals
The utilization of social and contact networks to address fundamental and applied questions about infectious disease transmission in wildlife is gaining increased attention. Epidemiological models that employ networks, with hosts represented as nodes and their interactions as edges, can provide much more accurate estimates of pathogen spread. Analysing network characteristics such as modularity (indicating network structure) can also offer valuable insights into how a pathogen may spread across a host population. However, the challenge of collecting network data often means that the utility of network modelling remains unrealized for most species.
This project aims to leverage recent advances in machine learning and spectral graph theory to construct models that can predict network characteristics serving as surrogates for disease spread for a species using trait data (e.g., total number of offspring). Our focus is on mammals, given that trait data is most comprehensive for this group. Utilizing trait-based proxies to gauge how effectively an infectious disease may spread in a population can provide a valuable starting point for understanding disease risk for mammals worldwide, including threatened species and species that are challenging to observe directly.
To some extent it is a follow-up on a previous Royal Society paper.
Meet The Team
Dr Matthew Silk
Research Fellow at the University of Edinburgh
"I am a quantitative ecologist who works at the interface of animal behaviour and disease ecology at the University of Edinburgh. My research mainly focuses on quantifying the role of social structure and dynamics on pathogen spread and maintenance. I have expertise in using social network approaches, with my research integrating empirical data collection, statistical modelling and theory across diverse taxa and ecological scales."
Dr Nick Fountain-Jones
Research Fellow at the University of Tasmania
"I am a disease ecologist at the University of Tasmania in the School of Natural Sciences. My research leveragess computational techniques to help untangle disease ecology in wildlife, humans and livestock. In particular, I utilize cutting-edge advances in network science, genomics and interpretable machine learning to help understand transmission and exposure risk as well as microbial assembly. The majority of my work involves viruses, but I work on a variety of other diseases from Lyme disease to Tasmanian devil facial tumour disease."