PRIORITIZING GENES IN RNA_SEQ EXPRESSION ANALYSIS USING THE CONSENSUS FROM MULTIPLE APPROACHES

Abstract

Gene expression data gives us the knowledge of total mRNA molecules in a given sample. It can be measured using various techniques: serial analysis of gene expression, northern blots, microarrays, Reverse-transcriptase polymerase chain reaction, expressed sequence tag, Ribonucleic acid Sequencing (RNA_Seq) technology, Massively parallel signature sequencing, etc. high throughput technologies provide a great revolution in this vision. RNA_Seq process gains importance due to its effective and cheap sequencing. This technology is greatly used by the researchers in genomics. The Cancer Genome Atlas (TCGA) has used this approach for tumor analysis. In case of RNA_Seq data, gene expression can be quantified using Reads per Kilo-base of exon model per Million mapped reads (RPKM). Now-a-days, Breast cancer is more prevalent in women’s causing to death. It is heterogeneous diseases, invasive or non-invasive in manner and categorized in hormone receptor-positive or triple-negative. The receptors can be human epidermal growth factors, hormone receptors (oestrogen and progesterone). In its signalling pathways, various gene are involved, to prioritize the gene for analysis by researchers various techniques are used. A very easy way to discovering interesting gene is comparison of expression profile of differentially expressed genes. Various approaches are available to extract the information from existing data using statistical methods. It can be Correlation coefficient method, Gene Rank, and Clustering. Correlation coefficient shows the linear relationship between two variables and their way of direction. It curtails the dimensionality of system. Gene Rank (GR) gives the ranking of gene in a given sample using Google Page Rank’s (PR) algorithm. Clustering tells the genes which are more correlated to each other comes under same cluster and different genes in different clusters, this separation can be done on the pattern similarity basis. From this, we found that all three techniques in combination can be used to make a decision for predicting the gene priority and can be used for further analytical advancements.