ClUSTERING METHODS IMPLEMENTED INTO MICROARRAYTOOL PROGRAM FOR ANALYSIS OF DNA MICROARRAY DATA

Abstract

Microarray technologies (DNA chips) allow to perform a quantitative analysis of expression of ten thousands genes. In this work a novel Microarraytool program was developed which allows to perform the cluster analysis and to compare the different experiments data by statistical analysis. Several clustering algorithms have been implemented into Microarraytool program: hierarchical clustering, k-means clustering, self-organizing maps (SOM) algorithm and self-organizing tree maps (SOTA) algorithm. The testing of these algorithms was performed using the Stanford Microarray Database for expression of 8613 individual genes in human fibroblasts after stimulation. The testing procedure revealed a correct performance of these algorithms implemented into Microarraytool program.