Neurodevelopmental consequences of prenatal alcohol exposure: a population based MRI study

Rationale
There is evidence that heavy alcohol intake during pregnancy disrupts normal brain development(1). However, it remains uncertain whether occasional use of moderate quantities of alcohol is harmful for the developing brain, since most recent epidemiological studies fail to find adverse effects(2). Nevertheless, evidence is emerging from different study designs more robust to biases including confounding, suggesting that even small amounts of alcohol during pregnancy could potentially affect neurodevelopment and result in lower cognitive and academic scores(3), as well as from animal models(4). Epigenetic marks (e.g. DNA methylation changes) resulting in altered regulation of gene expression provide one likely mechanism for the neurodevelopmental effects of alcohol. To our knowledge, the subtle effects of alcohol use on the brain have not been studied yet in large-scale population samples, nor has the potential mediating role of DNA methylation on such effects.

Aims & Objectives
The aim is to explore the associations of alcohol use during pregnancy with offspring structural brain morphology, with emphasis on their causal nature and the role played by DNA methylation. Objectives are:
1. To study the effects of prenatal exposure to moderate levels of alcohol on offspring structural brain morphology.
2. To explore the causal nature of the associations using causal analysis methods.
3. To investigate if DNA methylation is a causal mediator of the above associations.

Methods
Generation R, the Avon Longitudinal Study of Parents and Children (ALSPAC) and the Saguenay youth study will be used in this phd. Each of these cohort studies has available data on exposure and outcome measures and possible confounding and mediating factors (prenatal/postnatal environment, psycho-cognitive assessments, genome-wide genetic scans,DNA methylation data). These studies provide enough statistical power and design differences to yield methodologically robust results. The mediating role of DNA methylation will be interrogated using two-step Mendelian randomization(6).
The student will be encouraged to develop a work plan around their interests and will have the opportunity to familiarise with different research methodologies, including epigenomics and neuroimaging data processing and analyses(5), and causal analysis methods for epidemiology such as cross-cohort comparisons, negative control methods, and Mendelian randomization(6).