Evaluation of ordering methods for DNA sequence design based on ant colony system

Abstract

Hybridization between a DNA sequence and its base-pairing complement is crucial in DNA computing to retrieve the information stored in DNA sequences and operate a computation processes. Therefore, much works have focused on designing the DNA sequences for a reliable molecular computation. In this paper, Ant Colony System (ACS) is proposed to solve the DNA sequence design problem. ACS, which is based on Ant Colony Optimization (ACO) is an improvement of Ant System (AS) that uses some agents to obtain the solutions based on the pheromone in their colony. The DNA sequence design problem is modeled by four nodes, representing four DNA bases (A, T, C, and G) using the nearest-neighbor thermodynamic parameter's Watson-Crick base-pair □G°37 as distances between one node to other nodes. Seven ordering methods for ACS are presented in this study in order to obtain the best set solution. The performance of each of those methods are compared and evaluated to decide the best ordering method for this application. © 2008 IEEE.