CSS

Location: Building 8, Room 48: CS Conference Room

Starts: 12:00:00

Ends: 13:00:00

We propose the idea of using the topology of gene coexpression networks to characterise a specific phenotype or condition. Network based approaches are particularly useful when analysing human conditions/disease that are a result of the interactions among a large number of genes. We applied such a method to study Alzheimer's disease (AD) which is a progressive neurodegenerative disorder of the brain with good success. AD initially affects the entorhinal cortex (EC) and gradually spreads to other brain regions. Identifying the differences among affected brain regions may shed more light onto the disease progression/severity. We analysed microarray data of four brain regions - entorhinal cortex (EC), hippocampus (HIP), posterior cingulate cortex (PCC) and middle temporal gyrus (MTG) from AD affected and normal subjects. The 'activity' of a gene is represented by its connections in the coexpression network. Genes with zero topological overlap (neighbourhood analysis) between two region-specific networks were used to characterise the differences between regions. Our results indicate that MTG shows early AD pathology compared to the EC, HIP and PCC. We postulate that if the MTG gets affected later in the disease progression, post-mortem analyses of individuals with end-stage AD will show signs of early AD pathogenesis in the MTG, while the EC, HIP and PCC will have severe AD pathology. Such knowledge is useful for data collection in clinical AD studies where sample selection is critical and a limiting factor. Furthermore, identifying the gene expression differences among the affected brain regions using network topology may shed more light onto the disease progression.

Currently, we are extending this analysis to investigate the effects of low dose ionising radiation, < or =1-10 cGy, (LDIR) on human cells. Studies have indicated that extremely low doses of ionizing radiation can alleviate the adverse effects of subsequent higher doses. Investigating which biological processes are induced at low doses and whether certain genes have a dosage-based pleiotropic effect, is useful for the purpose of advancing the knowledge on human response to LDIR in order to assess the risks due to such exposures. Our data consists of the gene expression of cells exposed to 10 or 100 cGy measured at 0, 3, 8 and 24 hours. Results from such an analysis will also have implications in cancer and Alzheimer's disease research, as well as motivate systems biology approaches for understanding human conditions/diseases