The topological consequences of differential transcription in obese adipose tissue is key to our understanding of obesity beyond single genes or biological pathways, and may help identify network-based therapeutic targets. However, there is a dearth of network studies in obesity, especially for non-European cohorts. Here we report on an in-depth topological analysis of obesity-associated gene networks in Chinese subjects.
RNA sequencing was performed on visceral adipose tissue from 26 obese and 11 normal weight subjects (BMI 42.7+7.4 and 22.2+1.9, respectively). Obesity-associated differentially expressed genes were used to develop a gene coexpression network via GeneNet. Centiscape based network topological analysis predicted genes with large influence on network function. The coexpression network was modularized via MCODE and resulting clusters investigated for biological pathway enrichment via DAVID. A topological gene-set analysis identifying pathway subsections associated with obesity was conducted via Clipper. Finally, gene modules displaying differential correlation to obesity were identified via DICER and further analyzed for pathway enrichment.
461 genes were differentially expressed between obese and normal-weight samples (absolute logFC >1, FDR<5%). Several structural genes (IGHD, SGCA, SOWAHA, etc.) scored high on multiple network centrality measures. Coexpression network clusters were enriched in immune and mitochondrial respiratory chain related pathways (FDR<5%).Clipper analysis identified alternate signal paths in several pathways between obese and normal-weight subjects. DICER analysis revealed lower gene correlation modules in obese subjects. These modules were enriched for pathways related to oxidative phosphorylation, fatty acid metabolism and branched-chain amino acid metabolism.
These findings elucidate transcriptome-based topological signatures in obese Asians and identify key biological mechanisms altered in obesity-associated networks.