An Integrative Network Algorithm Identifies Age-Associated Differential Methylation Interactome Hotspots Targeting Stem-Cell Differentiation Pathways

James West,Stephan Beck,Xiangdong Wang,Andrew E. Teschendorff
DOI: https://doi.org/10.1038/srep01630
IF: 4.6
2013-01-01
Scientific Reports
Abstract:Epigenetic changes have been associated with ageing and cancer. Identifying and interpreting epigenetic changes associated with such phenotypes may benefit from integration with protein interactome models. We here develop and validate a novel integrative epigenome-interactome approach to identify differential methylation interactome hotspots associated with a phenotype of interest. We apply the algorithm to cancer and ageing, demonstrating the existence of hotspots associated with these phenotypes. Importantly, we discover tissue independent age-associated hotspots targeting stem-cell differentiation pathways, which we validate in independent DNA methylation data sets, encompassing over 1000 samples from different tissue types. We further show that these pathways would not have been discovered had we used a non-network based approach and that the use of the protein interaction network improves the overall robustness of the inference procedure. The proposed algorithm will be useful to any study seeking to identify interactome hotspots associated with common phenotypes.
What problem does this paper attempt to address?