Affective disorders are a leading cause of global disability, with heritability estimates ranging from 48 - 75% in major depressive disorder (MDD). Genetic studies have however struggled to identify consistent risk loci the predict risk of developing the disorder. One potential reason for this is due to the confounding impact of environmental factors, and also epigenetic changes that are not analysed in standard genetic studies. Childhood maltreatment is associated with persistent/recurrent depression, with evidence that the serotonin transporter gene (5HTT) interacts with childhood maltreatment to predict depression in population and clinical studies. Stress has also been associated with alterations in DNA methylation, specifically at specific loci in the 5HTTLPR. Through the investigation of methylation in the 5HTT, the aim of this study is to elucidate how the 5HTTLPR and exposure to childhood maltreatment might interact to increase risk of MDD. The sample includes 312 individuals with recurrent MDD (ICD10/DSM-IV) and 372 screened controls; the Child Trauma Questionnaire was completed to assess exposure to abuse and neglect, and demographic variables included as covariates. The 5HTTLPR was genotyped in the whole sample, and methylation of the 5HTT was assessed at one specific locus (cg22584138) and a second CpG-rich region of the promoter between -479 and -350 relative to the transcriptional start site using pyrosequencing. Methylation of these sites has previously been implicated in stress and abuse exposure (e.g.1 2 ). Differences in methylation between maltreatment exposure groups, genotype groups, and evidence for interaction will be described. This is the largest study to date investigating childhood maltreatment and 5HTTLPR methylation, and the first to investigate this in conjunction with depression status. Identifying the mechanism by which G-E interactions have their effect could provide novel interventions and pharmacological targets. This is particularly salient as a history of childhood maltreatment is associated with reduced pharmacotherapeutic response.