Codes within codes: How genetic variation influences disease through regional changes in methylation
Genetic differences that exist between individuals make a significant contribution to variation in disease risk across populations, but understanding the molecular path from genotype to phenotype is a big scientific question. For example, it is broadly understood that methylation (chemical modification to DNA) plays a central role in the rate of gene expression, which in turn influences disease outcomes. But elucidating such pathways requires novel statistical and computational methods applied to large datasets. New technology can now capture the genetic effects that influence methylation levels for specific genomic positions, but an enduring problem is that the rate of expression of a gene depends on the collective status of methylation levels at multiple sites across the region, only some of which are measured. One direction that this research can take is to understand how genetic variation influences regional methylation, and how this may have downstream effects on disease outcome.
Aims & Objectives
1. Develop methods for collapsing methylation information at multiple sites across a gene into a meaningful representation of regional regulation
2. Identify genetic influences for the regulation of regional methylation variation
3. Investigate the link between the genetic regulation of methylation, the regulation of gene expression by methylation, and the downstream effects on disease risk
The Avon Longitudinal Study of Parents and Children (ALSPAC) is a growing resource that has accumulated extensive phenotypic records on thousands of mothers and their children for over 20 years. Dense genotype data is available on a large proportion of participants (~18,000) and 1000 mother-child pairs have data on methylation levels in blood at 450,000 genomic positions at 5 time points as part of the ARIES project. Machine learning/feature selection and dimensionality reduction techniques will be applied to these data, followed by genetic analysis using genome wide association and whole genome methods. To test for causal influence of methylation on disease outcomes Mendelian randomization techniques will be applied either in this data set or by using more extensive data from large international collaborations.
Rakyan et al. (2014) Nature Reviews Genetics 12, 529-541
McRae et al. (2014) Genome Biology 15:R73
Smith and Hemani (2014) Human Molecular Genetics 15;23(R1):R89-R98