One way to view epigenomics is in terms of representing the software of living cells. It is increasingly recognised that complex diseases like cancer are not only driven by defects in the genetic machinery (i.e. the underlying hardware) but also by defects in the epigenome. However, to improve our understanding of how epigenomic aberrations may contribute to the causal development of diseases like cancer will require a systems-level epigenomics approach which integrates different omic data types together. In this chapter, we describe three systems-level statistical methods which have been successful in identifying novel biomarkers for ageing, for cancer risk and for early detection of cancer. In addition, these systems-level methods have provided us with substantial novel insights into systems-level aspects of carcinogenesis, which we also describe.