A novel parallel hybrid K-means-DE-ACO clustering approach for genomic clustering using MapReduce

Author

Bhavani, R. ; Sadasivam, G. Sudha ; Kumaran, Radhika

Author_Institution

Dept. of CSE, PSG Coll. of Technol., Coimbatore, India

fYear

2011

fDate

11-14 Dec. 2011

Firstpage

132

Lastpage

137

Abstract

The main aim of this paper is to design a scheme to identify the species from its genome sequence. Feature descriptors for a genome sequence are identified using MapReduce framework. Each feature descriptor is a three lettered keyword generated using A, T, C, G nucleotide bases. Genome sequences of related species are clustered by considering the feature descriptor count. MapReduce version of clustering model that uses K-means, Differential Evolution (DE) and Ant Colony Optimization (ACO) has been proposed. This MapReduce model improves accuracy as the entire genome sequence is considered. The inherent parallelism in the MapReduce model also enhances execution time efficiency.

Keywords

ant colony optimisation; biology computing; distributed processing; evolutionary computation; genomics; pattern clustering; A-T-C-G nucleotide bases; MapReduce framework; ant colony optimization; differential evolution; feature descriptors; genome sequence; genomic clustering; parallel hybrid k-means-DE-ACO clustering approach; species identification; Bioinformatics; Biological cells; Clustering algorithms; DNA; Feature extraction; Genomics; Vectors; Ant Colony Optimization; Differential Evolution; Genomic clustering; K-means clustering; MapReduce;

fLanguage

English

Publisher

ieee

Conference_Titel

Information and Communication Technologies (WICT), 2011 World Congress on

Conference_Location

Mumbai

Print_ISBN

978-1-4673-0127-5

Type

conf

DOI

10.1109/WICT.2011.6141231

Filename

6141231