Hot on the heels of personal genomics has come personal methylomics, where researchers use high-throughput methods to interrogate the DNA methylation landscape of individual human genomes. DNA methylation is thought to be an important mechanism by which cells regulate gene expression, but how does this regulation vary between and within individuals?
To appreciate how important a question this might be, imagine that a methylation pattern at a neurological condition risk allele is found to be specific to sufferers of that condition, or to those who will develop the condition later in life. In such a case, inter-individual variation may serve as a useful biomarker…
…but only if the methylation is not restricted to brain tissue (as biomarkers are generally most helpful pre-mortem), hence the importance of surveying intra-individual variation.
In this month's Genome Biology, Jon Mill and colleagues at King's College, London's Institute of Psychiatry report the analysis of matched post-mortem brain samples and pre-mortem blood samples. Obtaining data from matched samples enables the authors to consider – for the first time – how genome-wide methylation varies between blood and brain tissues within an individual. In addition, the authors surveryey a large cohort to shed light on inter-individual differences.
The King's College team (alongside their Baylor College of Medicine collaborators, led by Aleksandar Milosavljevic) achieved a fine-scale understanding of the brain methylome by mapping DNA methylation in several regions of the cortex, as well as in the cerebellum.
In an illuminating set of data, the Genome Biology study finds that DNA methylation for a given tissue is much more consistent between individuals than within a single individual (even within a single brain). But, despite this trend, inter-individual variation in brain tissue is replicated in the blood – a crucial observation. So can we remain hopeful that personal methylomics will pave the way for clinically practical biomarkers for neurological conditions? According to these exciting new results, the answer is yes.