Abstracts of the UK Molecular Epidemiology Group (MEG) Winter Meeting on The Future of Epidemiology: Biomarkers meet Populations. Newcastle University, United Kingdom. December 6, 2013

4. Sources of variation of white blood cell (WBC) DNA methylation in serial blood samples in the Breakthrough Generations Study (BGS) Montserrat Garcia-Closas Institute of Cancer Research, London, UK. Numerous epigenome-wide association studies (EWAS) are currently being performed on WBC DNA to identify associations between DNA methylation and disease risk. Strong correlations between WBC DNA methylation and environmental, lifestyle and other risk factors have also been reported. However, it is crucial to understand the stability of these methylation markers over time to determine whether case-control studies will need serial samples or whether single samples are sufficient. We estimated the intraclass correlation coefficient (ICC) for each probe on the Illumina 450K methylation array in paired samples collected ~6 years apart from 91 participants in the Breakthrough Generations Cohort, and evaluated relationships with reproductive and behavioural exposures. Approximately 20% of probes on the 450K array are variable between individuals and stable over a six year gap (ICC>0.50; stable variable methylated regions, VMRs). Stable-VMRs were enriched at approximately 1.2kb downstream from the transcription start site in the transition between the unmethylated promoter and methylated gene body. These stable-VMRs represent good candidates for EWAS using a single blood sample per subject. We also found that weight changes were related to changes in methylation levels, and that methylation levels in AHHR and F2RL3 genes (previously associated with smoking status) were associated with time since last smoked in former smokers. These findings provide support for further EWAS to identify biomarkers of exposure and disease risk. 5. Using epigenetics in epidemiology Caroline L. Relton University of Bristol, Bristol, UK; Newcastle University, Newcastle upon Tyne, UK. Epigenetics is predicted to become the focus of major advances in understanding the determinants and development of human disease. Epidemiological approaches can contribute to these advances by adopting strategies used in both conventional observational epidemiology and genetic epidemiology. The discipline of epidemiology aims to identify risk factors and identify targets for prediction, prevention and treatment. The epigenome is juxtaposed between the genome and the environment. There is a rapidly increasing body of evidence linking a variety of environmental and lifestyle factors to epigenetic variation. There is also an increasing recognition that epigenetic patterns are in part determined by underlying sequence variation. It is widely proposed that epigenetic variation lies on the causal pathway to disease and disease related traits. However it is equally important to note that epigenetic variation might be robustly linked to biomarkers or established risk factors that are not necessarily causally related to disease but may still serve as informative predictive tools. A range of epidemiological study designs and analytical strategies can be adopted, the choice of which depends upon whether epigenetic marks are considered simply as a biomarker of exposure, or to causally influence disease pathogenesis. 6. Using OMICS in human studies – investigations in the colorectal mucosa John C. Mathers Newcastle University, Newcastle upon Tyne, UK. The functional unit of the colorectal mucosa is the crypt populated by stem cells at the base of each crypt. Progeny from these stem cells differentiate to produce several lineages of columnar epithelial cells which form the single cell thick barrier between the gut lumen and the rest of the body. We have undertaken a series of studies using proteomics, gene expression and epigenetics approaches to identify early molecular changes which are associated with increased neoplastic risk and which may be modified by exposures such as diet and nutritional status and by age. We have observed widespread changes in the proteome of the apparently normal mucosa in mucosal biopsies [1] and changes in the methylation of the promoters of cancer-related genes [2, 3] from those at higher bowel cancer risk. In addition, nutritional factors and age appear to modulate these epigenetic marks [4]. Challenges in this work include the multi-cellular nature of the tissue which limits mechanistic interpretation of findings and the paucity of intervention study data which inhibits the drawing of causal inferences.