Determination Of Genetic Network From Time Series Gene Expression Data- A Modified Approach

Abstract

Genetic Network is one of the most revolutionary discoveries in the field of Genetic Engineering. Gene regulatory networks control biological functions by regulating the level of gene expression. Discovering and understanding the complex causal relationships within gene networks has become a major issue in systems biology, computational biology and bioinformatics. The benefits of characterizing gene interaction are many, for example, Genetic networks provide knowledge about functional pathway in a given cell, representing processes such as metabolism, gene regulation, transport, and signal transduction , the effects of drugs on a regulatory pathway can be found, the development of cancer in a cell can be tracked, etc. Genes are the building blocks of a body. Genetic code directs functional property of every living organism. Genes directly encode proteins that make up the cell to function properly. At first DNA is converted into a mature messenger RNA (mRNA). Then mRNA is read and converted into amino acid sequence. The information contained in the nucleotide sequence is read as three –letter word called codon. Now amino acids coded by codon together form a polypeptide chain that is later folded into protein. Few proteins are parked into promoter region of another protein and performs various jobs like turn it on or off, regulate the protein production rate etc. Thus we can say that each gene here is responsible for influencing other gene or it might influence itself. For this reason expression level of the working genes always changes with time. DNA microarray experiments today allow to monitor the output of gene regulatory networks by measuring the gene expression levels of thousands of genes. Our primary focus on this paper is to find out methods for finding out those sets of genes that have some contribution for the growth of a bacteria called ‘Burkholderia Pseudomalli’. At various phases of the growth of ‘Burkholderia Pseudomalli’ we performed computation using Microarray gene expression time series dataset. The dataset was obtained from GEO data base of NCBI website. Initially dataset contained information about 5289 genes in 47 consecutive time. iii The entire work was divided into two phases. The first phase was data reduction as performing computation with this huge sizes of the microarray data is a pressure hardware of the computers as well as it is very much time consuming. So a data reduction methodology was applied which finds out the responsible genes actively taking part in the overall bacterial growth process or we can say that the dominant genes responsible for the growth was found out. The second phase was formation of genetic network from genetic dependencies in various time series. Finally genetic network was of those genes that are responsible for the growth.Once genetic network was determined this network can be used to study various unknown biological process, metabolic pathway engineering, drug discovery etc.