Statistical Analysis of Type II Diabetes Gene Expression Datasets with reference to various Genetic and Environmental Factors using Open Source Software Project Bioconductor

Abstract

This doctoral study was designed to unravel various cellular mechanisms implicated in T2D pathology by robust statistical analysis of microarray datasets. Computational analysis of these datasets was carried out using R based Open Source Software project Bioconductor. Functional analysis of extracted genes generated from BioC analysis, was carried out using another Open Source Software Cytoscape. Besides analysis of individual microarray dataset, statistical meta-analysis of microarray datasets was also conducted as meta analysis. newline1. Statistical meta analysis of multiple datasets from different tissues, newline2. Re analysis of a single microarray dataset liver samples, newline3. Meta analysis of adipose tissue datasets using the Network Biology approach, newline4. Combined Differential Expression DE analysis of microarray datasets from different locations of adipose tissues like limb, visceral, and subcutaneous belonging to a common geographical location Asian Indians, newline5. Gene expression correlation analysis with clinical variables Insulin, HOMAR, HOMAB, TNF of the same datasets. newline. newlineAll the bioinformatics analysis eventually pointed to an understanding that it is inflammation of adipose tissue that sets the stage for development of Type 2 Diabetes. Various co morbidities associated with diabetes like diabetes nephropathy, limb amputation, pancreatic cancer, ovarian cancer, and dyslipidemia etc. are due to shared genes that causes extensive cross talk among cellular pathways. A fact that Asian Indians are genetically predisposed to sick fat also comes into consideration by the last two studies. newlineFinally this doctoral thesis endorsed that future direction toward development of therapeutics and diagnostics must embrace the concept of adiposopathy as the principal factor in the development of diabetes. newline newline