A modified computational model of ant colony system in DNA sequence design

Abstract

Major principle behind the development of computational intelligence is to address complex problem of real world application. Over the years, numerous computational intelligence algorithms have been developed in finding a solution to combinatorial optimization problem. Ant colony system (ACS) algorithm is one of the biologically inspired algorithms that have been applied to effectively solve various combinatorial optimization problems. In this study, ACS is going to be employed in solving DNA sequence design which is a study under the topics of DNA computing. The dependability of DNA computation is highly influenced by the information represents on the DNA strand and the strand reaction. We desire a set of stable double stranded DNA to retrieve the information encoded on the DNA sequence and to operate the computation without output error. To accomplish this, the DNA sequence design problem requires a set of objectives to be optimized and some constraints to be fulfilled. Therefore, DNA sequence design can be regarded as a constrained multi-objectives design problem. The multi-objective design problem is simplified into single-objective using the weighted sum method and objective functions used to obtain a good DNA sequence are Hmeasure, similarity, hairpin, and continuity. The sequence is subjected to two constraints which are Tm and GCcontent. The problem is modeled using finite state machine where each node represents the DNA bases {A, C, T, G}. In this study, 9 sets of studies have been conducted using 5, 7, 10, 15, 20, 25, 30, 35 and 40 agents/ants each with 100 independent runs. The number of iterations is set to be 300 for each set. Observation and analysis of the model with increasing number of ants was made and the performance of the model is measured by comparing the result with existing algorithm such as Genetic Algorithm (GA), Multi-Objective Evolutionary Algorithm (MOEA), Particle Swarm Optimization (PSO) etc. Based on the result, the suitable number of ants used for DNA sequence design was also proposed.