Determining Potential Yeast Longevity Genes via PPI Networks and Microarray Data Clustering Analysis

Abstract

Identification of genes involved in lifespan extension is a pre-requisite for studying aging and age-dependent diseases. So far, very few genes have been identified that relate to longevity. The process of analyzing each single gene one at a time can be a very long and expensive process. It is known that approximately 10% of 6000 yeast genes are lifespan related genes, however, less than 100 genes are identified as longevity genes. The interconnection of multiple genes and the time-dependent protein-protein interactions make researchers use systems biology as a first tool to predict genes potentially involved in aging. In this study, we combined analyses of protein-protein interaction data and micro array data to predict longevity genes. A dataset of all 6000 yeast genes was utilized and a protein-protein interaction ratio was used to narrow the dataset. Next, a hierarchical clustering algorithm was created to group the resulting data. From these clusters, conclusion of 6 highly possible longevity genes was drawn based on the amount of longevity genes in each cluster. Based on our latest information, one of our predicted genes is identified as a longevity gene. Wet lab experiments are applied to our predicted genes for supporting the findings.