DNA fragment assembly using optimization

Abstract

The DNA fragment assembly is an important phase required to obtain complete genomes. Optimization using nature inspired algorithms has been proposed by several authors. We present another nature inspired algorithm based on Particle Swarm Optimization and Differential Evolution. These algorithms are compared using a set of common benchmarks and showing some advantages in our proposed algorithm. We also applied the Traveling Salesman Problem (TSP) with better results than the nature inspired algorithms as we could obtain the true optima for 16 commonly used benchmarks for the first time to the best of our knowledge. The benchmarks are much smaller than the real organism assembly problems and scaling up from the benchmarks to real organisms presents important challenges. We propose a way to solve the scale up problems and test them using the Staphylococcus aureus COL Main Chromosome with the TSP approach.