SGER: Evolutionary Algorithms for Pathogen Defense

University Of Vermont & State Agricultural College, Burlington VT

Investigators

Abstract

Evolutionary computation uses principles from biological evolution to find optimal solutions to problems in engineering and computer science. Evolutionary algorithms are ideal for studying natural selection and optimization of the myriad of physiological processes that occur in organisms. In contrast, many other current modeling methods are limited in their ability to include dynamic, multi-objective optimization processes and results may be biased by initial parameter settings. Parameters in evolutionary algorithms can self-adapt, mimicking the evolution of processes such as mating system, migration, recombination and mutation. This research will use evolutionary algorithms to study multi-objective evolution of pathogen defense and local adaptation. An experiment using Tribolium flour beetles will validate the model and demonstrate the synergy between the computational efforts and empirical work. The project has broader impacts including stimulating an emerging research area through the novel application of recent developments in computer science and engineering to evolutionary physiology. Undergraduate, graduate and post-doctoral students will be trained in computational biology and the evolutionary physiology of host-pathogen interactions. The models will demonstrate the utility of evolutionary algorithms for modeling multiple defenses, self-adaptation, and spatial and temporal variation in pathogen prevalence. The topic of pathogen defense is a timely problem in both basic and applied biology, medicine and agriculture. These studies are relevant to genetically modified foods, and changes in pathogen abundance caused by global warming, as well as emerging infectious disease and drug resistance.