A High-Speed Two Dimensional Hierarchical Clustering of Microarray Gene Expression Data

Abstract

DNA micro array technology has become the most extensively used functional genomics approach in the bioinformatics field after genome sequencing. Revealing the patterns concealed in gene expression data offers a fabulous opportunity for an enhanced understanding of functional genomics. However, the large number of genes and the difficulty of biological networks greatly increase the challenges of comprehending and interpreting the resulting mass of data, which often consists of millions of measurements. The first step to address this challenge is the use of clustering techniques. Many clustering methods have been devised and used in the analysis of micro array data but less effort has gone into algorithmic speed up of those methods. In this research, quad tree based high-speed two dimensional hierarchical clustering is presented. In the hierarchical clustering process, the construction of the closest pair data structure in each level is the important time factor which determines the processing time of clustering. The proposed high-speed two dimensional clustering process uses the quad tree based data structure for finding the closest pair elements and thus reduces the processing time effectively and produces the better analysis of gene expression data.