Alexandre Bureau

CONTEXT: Our genome adapts to environmental influences, in part through epigenetic mechanisms, including DNA methylation. Variations in the quality of the early environment are associated with alterations in DNA methylation in rodents, and recent data suggest similar processes in humans in response to early-life adversity. OBJECTIVE: To determine genome-wide DNA methylation alterations induced by early-life trauma. DESIGN: Genome-wide study of promoter methylation in individuals with severe abuse during childhood. PATIENTS, SETTING, AND MAIN OUTCOME MEASURES: Promoter DNA methylation levels were profiled using methylated DNA immunoprecipitation followed by microarray hybridization in hippocampal tissue from 41 French-Canadian men (25 with a history of severe childhood abuse and 16 control subjects). Methylation profiles were compared with corresponding genome-wide gene expression profiles obtained by messenger RNA microarrays. Methylation differences between groups were validated on neuronal and nonneuronal DNA fractions isolated by fluorescence-assisted cell sorting. Functional consequences of site-specific promoter methylation were assessed by luciferase assays. RESULTS: We identified 362 differentially methylated promoters in individuals with a history of abuse compared with controls. Among these promoters, 248 showed hypermethylation and 114 demonstrated hypomethylation. Validation and site-specific quantification of DNA methylation in the 5 most hypermethylated gene promoters indicated that methylation differences occurred mainly in the neuronal cellular fraction. Genes involved in cellular/neuronal plasticity were among the most significantly differentially methylated, and, among these, Alsin (ALS2) was the most significant finding. Methylated ALS2 constructs mimicking the methylation state in samples from abused suicide completers showed decreased promoter transcriptional activity associated with decreased hippocampal expression of ALS2 variants. CONCLUSION: Childhood adversity is associated with epigenetic alterations in the promoters of several genes in hippocampal neurons.

There has been a great interest and a few successes in the identification of complex disease susceptibility genes in recent years. Association studies, where a large number of single-nucleotide polymorphisms (SNPs) are typed in a sample of cases and controls to determine which genes are associated with a specific disease, provide a powerful approach for complex disease gene mapping. Genes of interest in those studies may contain large numbers of SNPs that classical statistical methods cannot handle simultaneously without requiring prohibitively large sample sizes. By contrast, high-dimensional nonparametric methods thrive on large numbers of predictors. This work explores the application of one such method, random forests, to the problem of identifying SNPs predictive of the phenotype in the case-control study design. A random forest is a collection of classification trees grown on bootstrap samples of observations, using a random subset of predictors to define the best split at each node. The observations left out of the bootstrap samples are used to estimate prediction error. The importance of a predictor is quantified by the increase in misclassification occurring when the values of the predictor are randomly permuted. We extend the concept of importance to pairs of predictors, to capture joint effects, and we explore the behavior of importance measures over a range of two-locus disease models in the presence of a varying number of SNPs unassociated with the phenotype. We illustrate the application of random forests with a data set of asthma cases and unaffected controls genotyped at 42 SNPs in ADAM33, a previously identified asthma susceptibility gene. SNPs and SNP pairs highly associated with asthma tend to have the highest importance index value, but predictive importance and association do not always coincide.

BACKGROUND: It is unclear whether the association between impulsive-aggressive behaviours and suicide exists across different ages. METHOD: Via psychological autopsy, we examined a total of 645 subjects aged 11-87 years who died by suicide. Proxy-based interviews were conducted using the SCID-I & SCID-II or K-SADS interviews and a series of behavioural and personality-trait assessments. Secondarily, 246 living controls were similarly assessed. RESULTS: Higher levels of impulsivity, lifetime history of aggression, and novelty seeking were associated with younger age of death by suicide, while increasing levels of harm avoidance were associated with increasing age of suicide. This effect was observed after accounting for age-related psychopathology (current and lifetime depressive disorders, lifetime anxiety disorders, current and lifetime substance abuse disorders, psychotic disorders and cluster B personality disorders). Age effects were not due to the characteristics of informants, and such effects were not observed among living controls. When directly controlling for major psychopathology, the interaction between age, levels of impulsivity, aggression and novelty seeking predicted suicide status while controlling for the independent contributions of age and these traits. CONCLUSIONS: Higher levels of impulsive-aggressive traits play a greater role in suicide occurring among younger individuals, with decreasing importance with increasing age.