Genetic algorithm based clustering for gene-gene interaction in episodic memory. (7th July 2019)

• Record Type: Journal Article
• Authors: Tripathi, Sudhakar; Mishra, Ravi Bhushan; Sharma, Anand Kumar
• Is Part Of: International journal of bioinformatics research and applications. Volume 15:Number 3(2019)
• Page Start: 254
• Page End: 271
• Publication Date: 2019-07-07
• Subjects: clustering -- gene-gene interaction -- genetic algorithm -- hierarchical -- K-means -- SOM -- STRING 91; Bioinformatics -- Periodicals; 570.285
• Languages: English
• ISSNs: 1744-5485

Abstract

After the identification of several disease-associated polymorphisms by genome-wide association analysis, it is now clear that gene-gene interactions are fundamental mechanisms for the development of complex diseases. In this paper, we propose a genetic algorithm (GA) based clustering algorithm to identify groups of related genes in episodic memory. This clustering method required number of clusters and number of genes in each cluster and fitness function. In this paper, we have taken STRING 9.1 clustering method result on episodic memory. We have used “interaction between clusters” as a fitness function for the GA and have compared the result of GA based clustering algorithm with standard K-means, STRING 9.1 K-means, hierarchical and self-organising maps. We have evaluated the performance of all the above methods using Rand index, Jaccard index and Minkowski index. Our comparative study demonstrates that the proposed GA is close to hierarchical clustering method as far as the performance is concerned.