Using Network Structure to Analyze Vulnerability of French Pastures to Spread of Disease
It is no surprise to many people that most livestock animals are raised on pastures containing high densities of animals. Aside from “mad cow disease”, which is a disease passed to humans from eating contaminated meat, many other diseases threaten livestock and humans. Studying the spread of disease affecting livestock is important to keep both humans and animals safe. The aim of the study entitled “Analysis of the Spatial Organization of Pastures as a Contact Network, Implications for Potential Disease Spread and Biosecurity in Livestock” (Palisson, et. al) sought to use networks and graph theory to map out the vulnerability of pastures in France to the spread of livestock diseases.
In this study, the more than three million pastures in France were represented by nodes. An edge connected two pastures when their distance was below a set threshold, which was changed multiple times in the study to simulate different environments. For example, the researchers observed that the network had multiple small components when edges were only formed between pastures less than 1.5 meters apart. Thus, for diseases that required direct-transmission, it would be relatively difficult for such a disease to spread throughout all of the pastures in France. The largest connected component when the edge length was a maximum of 1.5 meters was only 0.1% of all French pastures. However, when the maximum edge length was increased to 500 meters, which would be the feasible transmission path of a vector-borne disease, the results were more concerning. There was one large component in the graph, and about 97% nodes were connected to the large component. Additionally, the researchers found that the average path length in this graph was about 296. The researchers also computed data for the clustering coefficients for graphs as well as how factoring in the ownership of pastures could change the resulting graph.
The results of the study have huge implications, both in the realm of networks and at the world at large. This study shows the feasibility and the flexibility of using networks and graph theory to study real-world scenarios. Additionally, this study exposed critical flaws in the biosecurity of France and the world at large. Vector-borne diseases could quickly transmit throughout the network and reach virtually every single pasture in France. The researchers were also able to apply network theory to develop potential solutions to the vulnerability they discovered. Some potential solutions they discussed included keeping animals indoors, making fences that prevent contact between pastures and using pastures that do not have neighbors. Additionally, the researchers in this study treated French pastures as their own isolated graph, unable to be connected to the pastures other countries. However, in the real-world disease could easily spread across country borders into Spain, Italy, and Belgium. This paper excelled in using network structure and theory to map a network of French pastures and discussed future areas of study and application as well.
Link: https://www.ncbi.nlm.nih.gov/pubmed/28060913