Title: Analyzing co-clustered and functionally connected human gene pairs
The order of genes (i.e. physical location of genes) within a eukaryotic genome is not random, rather they are because of several different kinds of evolutionary genomic rearrangements such as insertion and deletion etc. Several studies in the past – which focused on a few examples – have shown that a set of genes involved within a biological function are co-expressed and co-clustered together within the genome . In humans, some genes from the Purine metabolism pathway such as PAICS (GeneID 10606, encoding amidophosphoribosyltransferase EC 220.127.116.11) and PPAT (GeneID 5471, encoding SAICAR synthetase EC 18.104.22.168) are co-clustered in the genome at location 4q12, whereas other genes from the same pathway are located in different positions . My research focuses on examining the evolutionary association between gene order and biological functions. More specifically, I am interested in addressing the following fundamental research questions: a) How are co-clustered genes functionally and evolutionarily related? b) Why do genes from only a specific set of biological functions (and how many of them?) are grouped together within the human genome, and c) How do they affect metabolism, on a genome-scale. The gene-function association will be taken from Molecular Signatures Database (MSigDB), which has annotations from a range of databases. Genomic location from the human genome, and genomic functional association will be used to reconstruct a functionally co-clustered gene network. The network will be divided into topological modules and their functional association will be examined. The differential expressions will be studied using limma package in R. For a pair of genes, the differential expression pattern across conditions will be examined and, using statistical analysis, a gene co-expression network will be constructed. The resulting network will be used to address the objectives of the study.