Whole-genome analysis of neural epigenomes

Research over the past decade has demonstrated the influence of environmental conditions on the neural epigenome. Such effects are apparent at the level of DNA methylation and histone post-translational modifications (PTMs), and occur in response to environmental signals operating during early development or in adulthood. These epigenetic modifications then mediate the influence of environmental conditions on genomic transcription and complex phenotypes. Two major questions concern the first pathways by which environmental conditions produce epigenetic remodeling and second the extent of such effects across the genome.

Recent advances in molecular biology/genetics reveal that variation in DNA methylation often occurs either in alternative forms of methylation, such as 5-hydroxymethylation, or at non-CpG sites. The exact function of these variants in relation to transcription is not clear, not do we know of the relative sensitivity of such variants to environmental influences. Moreover it is not clear how modifications to DNA methylation, in whatever form, align with histone modifications. Our current studies use next generation sequencing with microdissected brain samples and defined cellular populations to define the impact of ‘clinically-relevant’ environmental conditions across the genome using rodent models. A clinically-relevant condition is defined as one that mimics conditions that predict mental health outcomes in humans. Examples of this approach are studies of variations in maternal care or environmental enrichment/social isolation. Moreover, the use of genetically modified rodent models allows us to examine the influence of genotype on the effect of environmental conditions on specific epigenetic signals.

Epigenetics: when the environment modifies the genes

We currently collaborate with the McGill Centre for Innovation, which is linked to the International Human Epigenome Consortium (IHEC) to use state-of-the-art sequencing approaches to the study of coordinated epigenetic states in multiple brain regions. These studies permit the concurrent analysis of multiple forms of DNA methylation, histone PTMs and transcription. Studies with multiple transgenic models will permit the study of essential signaling pathways that link environmental conditions to epigenetic states. The Ludmer Centre for Neuroinformatics and Mental Health offers state-of-the-art bioinformatics support for the analysis of the resulting whole-genome sequencing data.

Gene x environment interactions and the epigenome

We study the interactive effects of genetic and environmental factors using rodent models, as described above, as well as in human samples. We use genome-wide analyses of DNA methylation in biosamples derived from human subjects. These studies involve a range of collaborations that include intervention studies, such as the Nurse Family Partnership project, longitudinal birth cohort studies (such as the MAVAN project described below), clinical populations and high-risk cohorts (e.g., samples of individuals victimized by abuse in childhood). These samples also include genome-wide analysis of single nucleotide polymorphisms and copy number variants using various platforms. Most studies, especially the longitudinal birth cohort studies such as MAVAN or the Singapore-based Growing Up in Singapore Towards healthy Outcomes include extensive phenotyping over development as well as well as comprehensive measures of the social and physical environment and neuroimaging These data sets provide novel opportunities for data mining as well as for the development of novel computational approaches.