In Search of the Optimal Root System: Multi-Objective Optimization of Root Systems Using Evolutionary Algorithms.

Global agricultural productivity is limited by low nutrient availability. Selection of plants with optimal root architecture can improve nutrient acquisition. Root architecture is dependent on a number of traits. Evaluation of the optimal combination of root traits under varied environmental conditions is a daunting task. This problem can be thought of as a search for the optimal solution in a multidimensional decision space. In fields of study, ranging from aerospace engineering to gene expression profiling, multi-objective evolutionary algorithms (MOEAs) have been proven to be useful in searches across multidimensional search space optimizing multiple conflicting objectives. In this study we demonstrate the application of the BORG MOEA to find optimal root architectures for acqusition of N and P by maize and common bean. The three dimensional structural functional root architectural model, SimRoot, was linked to BORG MOEA and the optimization runs were evaluated for several generations of solutions to find the optimal root system in terms of biomass production, nutrient acquisition and root carbon costs.