Title: Environmental demogenetics of potato late blight in Colombia
Abstract: The potato late blight is known for its part in the Irish potato famine during the 19th century. This disease has been extensively studied ever since. However, despite being one of the most well studied plant diseases in the world it remains one of the biggest threats to global food security. The late blight is caused by the Oomycete Phytophthora infestans. This is an hemi-biotrophic pathogen that infects the economically important crops potato (Solanum tuberosum) and tomato (Solanum lycopersicum).
In Colombia due to the prevalence of this disease and the extended use of susceptible potato cultivars, the main control strategy against this disease is the continuous application of fungicides. This constitutes a major problem because the repeated exposure of P. infestans to these fungicides results in the development of acquired resistance, and the effects on the health of the growers and the increased costs of continuously using fungicides during the growth cycle of the crop.
One way of reducing the continuous use of fungicides that has been practiced with some degree of success is the use of simulation and epidemiological models. These models project the proliferation of P. infestans based on environmental conditions such as temperature and relative humidity. This allows the growers to optimize the use of fungicides by applying them exclusively during the most favorable periods for the late blight development instead of a continuous use.
These models however have a few limitations. First, these do not consider the spatial configuration beyond each individual field. This could be an important addition due to the possibility of P. infestans dispersal between fields. Second, these disregard other possible management strategies and conditions that could be informative for the projection of late blight. And third, these are deterministic mechanistic models which have required decades of study to find the response of P. infestans to a variety of environmental conditions. Even though there have been adaptations of these models to tropical conditions, these models have a limited applicability outside the US because of the differences in environmental conditions and the responses of the different lineages present elsewhere.
Our main goal in this work is to develop an integrative model that considers several sources of information including environmental, epidemiological, genetic and spatial within a Bayesian learning framework. The idea is to start with a simple model that can be calibrated through this approach using collected field data. This would allow us to develop both a model for potato late blight in Colombia which would be continually calibrated with newly collected information, and a generic framework that can be used to develop models for lesser known plant pathogens that could be calibrated relying heavily on the collected field data and not previous mechanistic studies.