Identifying Most Significant Genes on Imputed Gene Expression Dataset

Abstract

Due to advance research in the field of Microarray technology large amount of gene expressions are generated. These gene dataset helps in getting more insight of cells and their functioning. While capturing gene expressions noise get added in the dataset. For the accuracy of downstream analysis it is necessary to preprocess this data. This helps in accurately identifying most significant and co-expressing genes. In this paper, we have implemented USQR algorithm for data reduction after applying normalization and discretization. We have used serum dataset containing 1700 genes.