Chaos inspired Particle Swarm Optimization with Levy Flight for Genome Sequence Assembly
This work addresses genome sequence assembly for computational biology and personalized medicine, representing an incremental improvement over existing PSO methods.
The paper tackles the complex combinatorial optimization problem of genome sequence assembly by proposing a new variant of Particle Swarm Optimization (PSO) integrated with Chaos and Levy Flight to balance exploration and exploitation, achieving better performance, reliability, and consistency compared to other PSO variants on four DNA coverage datasets.
With the advent of Genome Sequencing, the field of Personalized Medicine has been revolutionized. From drug testing and studying diseases and mutations to clan genomics, studying the genome is required. However, genome sequence assembly is a very complex combinatorial optimization problem of computational biology. PSO is a popular meta-heuristic swarm intelligence optimization algorithm, used to solve combinatorial optimization problems. In this paper, we propose a new variant of PSO to address this permutation-optimization problem. PSO is integrated with the Chaos and Levy Flight (A random walk algorithm) to effectively balance the exploration and exploitation capability of the algorithm. Empirical experiments are conducted to evaluate the performance of the proposed method in comparison to the other variants of the PSO proposed in the literature. The analysis is conducted on four DNA coverage datasets. The conducted analysis demonstrates that the proposed model attain a better performance with better reliability and consistency in comparison to other competitive methods in all cases.