Abstract:
DNA microarrays have emerged as the premier tool for studying gene expression on a genomics scale. They provide a format for the simultaneous measurements of the expression levels of thousands of genes in a single hybridization array. Scientists seeking to harness the potential of this technique are challenged by the large quantities of data produced. For tracking, integrating, qualifying and ultimately deriving scientific insight from the experimental results, various
tools are required. In general, a well designed database, an interface for data
entry and query, an image analysis software, normalization and filtering routines
and software for data analysis and visualization are needed. For data analysis,
various statistical and machine learning techniques have been applied. The
main task is making groups of genes having the similar expression pattern. For
this, methods such as hierarchical clustering, k-means clustering, self organizing
maps, neural fretwork, support vector machine etc have been applied.
