Aims - We aim to associate OXTR methylation level with measures of autistic like traits in cord blood derived from 1000 boys from the ALSPAC study.

Hypotheses - We hypothesize that (1) a relationship between methylation and traits exists and (2) that this relationship may depend on rs53576 genotype.

Rationale - Methylation of DNA exists in all human cells. While tissue-specific variation in DNA methylation has been observed, recent work indicates that, on the whole, methylation patterns are relatively conserved across tissue types within individuals. Published work from Connelly identified increased DNA methylation of a regulatory region of OXTR in ASD temporal cortex derived from Brodmann's area 41/42 that resulted in concomitant decreased transcription of OXTR. This brain region, a portion of superior temporal sulcus (STS), has been strongly implicated in social perception by virtue of its role in biological motion perception and theory of mind. Two of these ASD-specific DNA methylation changes (CpG sites -934 and -860) were also apparent in the blood of ASD individuals, and unpublished data from the PI indicates that methylation of CpG site -934 may be heritable. Importantly, this region of OXTR is not present on the current arrays used in recent large scale, methylome-wide studies in ASD; thus it is necessary to assay this region using other methods for replication. We recently replicated these methylation changes in a separate sample of male ASD individuals and their unaffected male siblings derived from the Simons Simplex collection. These data suggest that 1) increased DNA methylation of OXTR may lead to reduced OXTR gene expression in the ASD brain, 2) DNA methylation increases track with the ASD phenotype and 3) peripheral blood can be used a biomarker of brain methylation.

Data from the Connely lab establish an important relationship between measures of OXTR DNA methylation in the blood and perception of animacy in dynamic displays. The attribution of social meaning to these displays is consistently disrupted in ASD. Among typically developing individuals, viewing these displays recruits brain regions associated with mentalizing processes, including STS, temporal poles, and medial prefrontal cortex. Given the likely mediating role of oxytocin in social perception, we predicted that individual differences in OXTR methylation might affect response in these brain regions when viewing animate motion in dynamic displays. We found that higher levels of OXTR methylation predicted greater activity in STS and cingulate gyrus. Our results indicate that OXTR methylation may impact the degree to which individuals are sensitive to displays of animate motion. This perceptual sensitivity could be indicative of a social style that varies within the population and is compromised in ASD. Preliminary data in a larger sample of healthy Caucasians suggests that OXTR methylation may interact with a common functional variant along OXTR (rs53576) to influence the social brain. The A allele of rs53576 is related to decreased psychological resources, dispositional empathy and stress reactivity, as well as to structural and functional differences in oxytocinergic brain sites. We examined the relationship between OXTR methylation and response in anterior cingulate cortex to social attribution and emotional faces and find an inverse correlation based on rs53576 genotype.

Sample selection - 1000 male study participants weighted by autism like traits will be selected

Exposure Variables - NONE

Outcome Variables - Variables that fall under the catagories of communicative, social, and repetitive behavior; as in Table S2 [Steer, Golding, and Bolton, PLOS ONE 2010].

Confounding Variables - Although not considered a core requirement for the diagnosis of ASD, many children exhibit other traits such as learning difficulties, specific language impairment (SLI), ADHD, ODD/ CD, anxiety problems and special education needs.

Material Requested - 200 ng of DNA at 10ng/ul for genetic and epigenetic assays.

Aims: To estimate the prevalence and correlates of tongue-ties in neonates in ALSPAC.

Aims: The aim of this project is to run growth trajectory analysis to examine the relationship between residential mobility and BMI/fat mass throughout childhood and adolescence. Growth trajectory analysis will permit differences in BMI/fat mass trajectories from key stages in life to be calculated, to examine critical periods of risk such as during schooling years (pre-school, primary, secondary).

Aim: To investigate the relationship between attention and pain in childhood.

Background: There is growing evidence of a link between pain and attention. Several studies have demonstrated that being in pain consumes our attention and reduces our effectiveness on current tasks (Moore, Keogh & Eccleston, 2012, 2013). Those with better attentional resources may also be better served by distractions from pain (Legrain, Van Damme, Eccleston, Davis, Seminowicz & Crombes, 2009; Verhoeven, Dick, Eccleston, Goubert & Crombez, 2012). In another strain of research, there is evidence that childhood intelligence, a concept closely related to attention, is related to various health outcomes such as life span (Whalley & Deary, 2001), late-onset dementia (Whalley et al, 2000), cardiovascular disease (Hart et al, 2004) and psychiatric disorders (Batty, Mortensen & Osler, 2005). The proposed research will extend and bridge these two areas of literature on cognition and health by investigating the relationship of chronic pain to intelligence and attention. This will extend our understanding of the relationship between a) childhood attention and experiences of pain and more broadly b) childhood intelligence and health outcomes.

Aims

The aims of this study are as follows:

1. To conduct a meta- and mega-analysis genome-wide association study of psychotic experiences in adolescents and young adults by combining data from multiple samples including ALSPAC. Further analyses will investigate the effects by chromosome location and variant class, and explore gene-gene and gene-environment interactions in the combined datasets.

2. To test whether genes influencing risk for adult psychiatric conditions such as schizophrenia and bipolar disorder and major depression (using polygenic risk scores) are shared with those for psychotic experiences in adolescence and young adulthood in a dataset of the combined samples including ALSPAC.

3. To conduct genome-wide complex trait analyses (GCTA) of psychotic experiences in adolescents and young adults in the combined dataset in order to estimate the measured heritability of these trait experiences.

Adiposity and blood pressure have been associated with derangements in cardiac structure and the retinal microvascular architecture in both adults and children. These derangements in adults including left ventricular (LV) hypertrophy (a consequence of increased left ventricular mass) and retinal diameters are associated with an increased risk of myocardial infarction and cardiovascular disease (CDV) mortality. To further our understanding of these associations, the link between the microvasculature and cardiac structure and to tease out the separate and synergistic impact of CVD risk factors, studies need to be undertaken in children and young adults assessing both the microvasculature and cardiac structure, before the disease process is established.

The measures of cardiac structure (left atrial size (LA size), relative wall thickness (RWT) and posterior wall thickness (PWT)) and the retinal microvasculature, provide surrogate measures of CVD which can be assessed non-invasively, are highly reproducible and track through childhood and early adulthood. The identification of a strong link between the microvasculature and measures of cardiac structure could potentially allow further risk stratification to identify sub groups at increased risk of coronary heart disease (CHD), who would not normally be identified by assessment of traditional risk factors. To date these studies have only been undertaken in adults and the results have been varied, a likely consequence of assessing populations with advanced disease, where confounding is difficult to fully adjust for. The Avon Longitudinal Study of Parents and Children (ALSPAC) has followed a cohort of 14,541 children from birth to late adolescence, with measures of adiposity, blood pressure, retinal microvascular architecture and cardiac structure. The ALSPAC study provides the ideal setting to further our understanding of the link between the microvasculature and cardiac structure.

Hypothesis: We hypothesised that retinal microvascular measures (potential predictors) would be associated with derangements in cardiac structure (outcome), independent of cardiovascular risk factors and body mass index (BMI) in adolescence.

Variables: The current study would include all children with cardiac images and ambulatory blood pressure acquired at the 17 year clinic and retinal images acquired at the 11+ year clinic. Other measures to be included in the analsyis are: Gestation, age, mother smoking during pregnancy, mother drank during pregnancy, BMI 17 year clinic, systolic and diastolic BP at the 17 year clinic, sex, paternal SES, height 17 year clinic, pubertal stage at measurement. With the exeption on the ambulatory BP data at the 17 year clinic I have all of the data required for the analysis.

The associations between retinal diameters (potential predictors) and cardiac measures (outcomes) will be modelled using multiple linear regression. The associations between retinal diameters and cardiac measures will be presented unadjusted, a second model including retinal diameter (separately for each retinal vascular measure) gestation and sex, and potential confounding factors (paternal SES, maternal smoking during 1st three months pregnancy, pubertal stage at measurement outcome and height and height2). A third model will include potential mediating factors BMI and systolic BP at age 17 years.

Specific aims are:

1. To study associations of abuse in childhood including sexual, parental physical and emotional abuse with cardiometabolic health in middle age (Framingham CVD risk score, atherosclerosis (cIMT, arterial distensibility), pulse wave velocity, adiposity, blood pressure, lipids, insulin, glucose, inflammatory markers).

2. To study associations of abuse in childhood including sexual, parental physical and emotional abuse with women's reproductive health across the lifecourse (age at menarche, menstrual regularity, time to conception, seeing a physician for possible infertility, pregnancy outcomes: pregnancy losses, preterm delivery, mode of delivery, birth weight, parity, age at menopause).

3. To describe longitudinal patterns of intimate partner physical and emotional cruelty throughout the life course, and of physical and emotional abuse of offspring, in men and women.

4. To compare maternal and paternal reports of IPV.

5. To examine patterns of IPV across two generations in the same families.

6. To examine associations of longitudinal patterns of intimate partner physical and emotional cruelty with cardiometabolic (men and women) and reproductive health (women).

7. To examine whether any associations identified in aims 1, 2 and 6 are mediated by established cardiometabolic risk factors such as smoking, alcohol consumption, eating disorders, adiposity, and mental health.

8. To examine whether any associations identified in aims 1, 2 and 6 are mediated by differential DNA methylation.

Foods are generally consumed in combination; therefore dietary recommendations should consider diet as a whole, rather than individual foods or nutrients. We know that dietary intake throughout the life course is involved in the development of lifestyle diseases, including cardiovascular disease (CVD) and obesity which are currently endemic in the UK. This project aims to provide an insight into nutritional life course exposures and the potential of these exposures to affect markers of CVD.

Studies have previously linked childhood obesity with CVD in adulthood (Lloyd et al., 2010) and therefore asfood behaviours established in childhood/adolescence may ultimately go on to affect adult cardiovascular health it is important to adopt a healthy lifestyle early in life in order to decrease later disease risk. Observing dietary patterns throughout the life course should be beneficial in calculating the time point at which nutritional intake may be most important and also whether tracking one type of dietary pattern over a period of time or changing to a different diet pattern renders an individual more/less likely to be at risk of disease. Dietary patterns are primarily derived via two statistical methods: cluster analysis (CA) and principal component analysis (PCA). Both of these methods have been found to give similar results in the ALSPAC study at 7 years of age (Smith et al., 2011).

Tracking over time is easier to quantify for patterns that have been derived using cluster analysis as this method assigns an individual to one category only at each timepoint. Change in category can then easily be determined. In comparison, PCA results in a score for each individual for each pattern obtained. We will therefore examined patterns obtained from CA in the first instance.Four clusters have been observed in ALSPAC using food diary data at 7, 10 and 13 years of age (Northstone et al., 2013). We will use this information to investigate whether dietary patterns and their tracking have any implication upon known risk factors for CVD (fat mass, blood pressure, CIMT and blood lipids) observed in the cohort at 15 and 17 years of age. Socioeconomic status is a major confounder for dietary intake (Northstone et al., 2012 & 2005). It is hypothesised that there will be a correlation between dietary patterns and measured risk factors for CVD, such that a more healthy pattern will infer decreased risk and that any associations may strengthen with pattern tracking (e.g. where an individual is consistently assigned to the same pattern over time).

Please note: This is a mini-project that Anna Guyatt (Wellcome 4 year phd student) and Bernice Knight (clinical lecturer) will be working on. DR has direct access and will liaise with Kate Northstone to anonymise the IDs (this has already been discussed with Kate).

Background and Aims: Autism spectrum disorders and associated traits are over-represented in males. A popular hypothesis to explain this gender bias is the 'extreme male brain' theory (EMB), which is an extension of the empathising-systemising theory of sex differences in cognitive styles (Baron-Cohen 2002; Baron- Cohen and Hammer 1997). It suggests that male brains are hardwired for the drive to understand and construct systems based on if-then rules (i.e., systemize) while the female brain is programmed for the drive to understand the mental state of others and experience appropriate reactions (i.e., empathizing). The EMB hypothesis suggests that beyond the biological sex, higher levels of fetal testosterone underlies the male brain (Auyeung and Baron-Cohen 2008). Baron-Cohen has suggested that the deficits observed in autism of extreme systemising and less empathising may suggest that higher exposure to testosterone during fetal life may be aetiologically linked to autistic traits. The ratio of the index finger (2nd digit) to the ring finger (4th digit), commonly referred to as 2D:4D is a commonly used putative marker of fetal testosterone relative to fetal estrogen activity. Although a number of studies have tried to test the relationship between 2D:4D and autism/traits, the samples have been relatively small and selected; and the results are inconclusive. There is a lack of population based studies.To address this gap in the literature, we aim to study the association between 2D:4D and autism and its component traits, testing the extreme male brain hypothesis.