Peterson, A. Townsend, et al. “Ecologic niche modeling and potential reservoirs for Chagas disease, Mexico.” (2002).
doi: 10.3201/eid0807.010454
Townsend et. al. applies ecological niche modeling to improve the understanding of epidemiologically important vectors and parasite-reservoirs of Chagas disease using the Neotoma (pack-rat) and Triatoma species affiliation as a study system. The purpose of the study was to determine potential risk areas with ENM using primary occurrence data of various triatomine species to identify the degree of host affiliation with Neotoma sp. (conventionally known to be a strong vector-host affiliation) within Mexico. The generated model output was compared with field observations to test the quality of the model. Both ecological niches and potential geographic distribution were generated using the Genetic Algorithm for Rule-Set Prediction with environmental/ecologic data coverage (11 conventionally used variables). From the data available of the various Triatoma species, species with small sample sizes were not used in niche model. Prediction output was compared with known distributions of both rodent and vector, using the percentage of overlap as species associations and potential disease transmission. Results: Predicted distribution performed well, and triatomines were indeed found at locations predicted from the model, which also overlapped rodent distribution, suggesting a strong affiliation with the host. However, the model did fail to predict some species overlap which has been observed in the field, and this may be due to small sample of the species. This result limited the reliability of model to predict disease risk.
Regardless, this paper suggests that ecological niche modeling and species distribution prediction with GARP is useful tool in determining potential interactions between disease vectors and reservoir hosts. Can also suggest evolutionary relationships between vector and host depending on the percentage of overlap and identify further species interactions that were not previously identified or view potential jumps from sylvatic affiliations to more peridomestic habitat types. Thus, this method can provide a useful supplementary tool to link current/future Chagas disease risk given host-vector distributions in addition to detecting shifts in host affiliation and distributional boundaries.