An evaluation of the ecological niche of Orf virus (Poxviridae): Challenges of distinguishing broad niches from no niches

Abstract

Contagious ecthyma is a skin disease, caused by Orf virus, creating great economic threats to livestock farming worldwide. Zoonotic potential of this disease has gained recent attention owing to the re-emergence of disease in several parts of the world. Increased public health concern emphasizes the need for a predictive understanding of the geographic distributional potential of Orf virus. Here, we mapped the current distribution using occurrence records, and estimated the ecological niche in both geographical and environmental spaces. Twenty modeling experiments, resulting from two- and three-partition models, were performed to choose the candidate models that best represent the geographic distributional potential of Orf virus. For all of our models, it was possible to reject the null hypothesis of predictive performance no better than random expectations. However, statistical significance must be accompanied by sufficiently good predictive performance if a model is to be useful. In our case, omission of known distribution of the virus was noticed in all Maxent models, indicating inferior quality of our models. This conclusion was further confirmed by the independent final evaluation, using occurrence records sourced from the Centre for Agriculture and Bioscience International. Minimum volume ellipsoid (MVE) models indicated the broad range of environmental conditions under which Orf virus infections are found. The excluded climatic conditions from MVEs could not be considered as unsuitable owing to the broad distribution of Orf virus. These results suggest two possibilities: that the niche models fail to identify niche limits that constrain the virus, or that the virus has no detectable niche, as it can be found throughout the geographic distributions of its hosts. This potential limitation of component-based pathogen-only ENMs is discussed in detail.