Locus 1 (15q14). We completed a GWAS for the phenotype 'refractive error' in our Discovery cohort, the population-based Rotterdam Study (N = 5,328 subjects, using SNPs genotyped with the Illumina Infinium II HumanHap550 chip v3.0 array) . The results were replicated in 4 additional cohorts (combined N = 10,280 individuals in the replication stage): the Rotterdam Study II cohort, the Rotterdam Study III cohort, the Erasmus Rucphen Family (ERF) Study cohort and the TwinsUK cohort. The findings have been reported [1].

Locus 2 (15q25). In collaboration with the TwinsUK group, a second region of strong association was observed at 15q25. This association was replicated in a combined sample comprising the above cohorts, along with subjects from the 1958 British Birth Cohort and the Australian Twin Eye Study cohort (combined N = 13,414 subjects). These findings have also been published [2].

We propose to carry out additional replication studies for 19 SNPs in these two regions of strong association in a number of cohorts of European ethnicity. The ALSPAC cohort is suitable because it is ethnically well-matched to our original study cohort and is also a population-based sample, not selected based on eye phenotypes. We expect this will aid the differentiation of causal variants and non-causal variants in the associated regions.

Replication analysis on the 19 SNPs will be performed by Dr Jez Guggenheim, who is part of the ALSPAC Myopia team, and summary results will be sent to Dr Virginie J.M. Verhoeven. To facilitate the replication step on the ALSPAC side, we ask for a DTA of the 19 SNPs to Dr Jez Guggenheim who will perform the work at Cardiff University, or in Bristol if required.

Prof Tim Cole is applying for funding to further develop the SITAR method of growth curve analysis. There are three distinct objectives: to extend the SITAR methodology, to work with collaborators to apply the method to their own existing data, and to develop a library in the R statistical language35 to make the method more accessible to other researchers. The use of ALSPAC data relates to objective 2, and complements work already planned as part of Laura Howe's MRC Fellowship.

Not available

Aims and objectives

The aim is to use a cross-cohort comparison between Pelotas and ALSPAC to examine the causal association of maternal age with adverse perinatal and later offspring outcomes. Specific objectives are:

Compare the magnitudes of association of SEP with maternal age, using relative indices of inequalities, between the four cohorts: threePelotascohorts and ALSPAC.

1. Compare the magnitudes of association of maternal age with birth weight and gestational age between the four cohorts: threePelotascohorts and ALSPAC.
2. Compare the magnitudes of association of maternal age with IQ and educational attainment between the four cohorts: three Pelotas cohorts and ALSPAC
3. Compare the magnitudes of association of maternal age with behavioural problems: three Pelotas cohorts and ALSPAC
4. Compare the magnitudes of association of maternal age with smoking and alcohol initiation and use between the four cohorts: three Pelotas cohorts and ALSPAC

This project is a continuation of research previously carried out using data from the National Survey of Sexual Attitudes and Lifestyles (Natsal) 2000. This research identified statistically significant associations between early sexual debut and risky or non-normative sexual behaviour later into adulthood (aged up to 29). Those respondents whose sexual debut occurred before the age of 16 had increased odds of engaging in certain risky or non-normative sexual activities later in life. This association maintained significance with the addition of potential confounders. This PhD aims to explore this phenomenon further.

An essential component of this research is to identity what factors influence contemporary young persons' engagement in sexual activities early on in life. While existing research provides evidence of socio-psychological and economic predictors (ranging from personality type, to family structure, parenting and enjoyment of school) such research in Britain is limited, particularly that of a longitudinal nature examining the multitude of influences which may cause, interact and feedback on one another. The majority of British studies in this topic area rely on cross-sectional data and concentrate on only a few predictors. Given the older age and the subject coverage of the British birth cohorts, longitudinal data with the potential to examine contemporary teenagers' trajectories into their sexual live is essentially non-existent, with the exception of the ALSPAC cohort.

Analyses will be carried out separately for males and females, so allowing identification of gender differences. Rather than stopping at the identification risk factors, further structural equation modelling and pathway analysis will be carried out in an effort to identify mediating factors. The longitudinal nature of the data lends itself particularly well to pathway analysis; once theoretical mediators are identified from the literature they can be analysed in chronological order.

In addition to the focus on age at sexual debut, this project also seeks to explore the concept of 'sexual competence' as defined by Wellings et al., 2001 as a sexual encounter characterised by the absence of duress and regret, autonomy of decision, and use of a reliable method of contraception. A similar composite measure as that used by Wellings et al can be constructed using ALSPAC data. This measure will undergo critical review in order to establish its validity before it is included in final analyses. Latent class analysis of sexual competence at each time point when the 'Romantic Relations' questions were completed will allow the exploration of how sexual competence changes or develops over time; this will of course be limited by how many people have actually begun engaging in sexual activity at a young age and will be subject to censoring. Using latent class analysis in this way will allow the identification of distinct trajectories of sexual competence, which can then be analysed against the relevant predictors.

Attention will also be paid to other 'risky' behaviours, such as smoking, alcohol consumption and drug use alongside sexual behaviour. The clustering of a variety of risk behaviours within certain individuals has been reported previously. The record of these behaviours in the ALSPAC cohort at different ages lends itself well for the exploration of behavioural trajectories; examining the temporal relationship between smoking, alcohol consumption, drug use and sexual activities in adolescence.

This project would seek to use the breadth of data available in ALSPAC which pertains to the influences on development of the children under study. Broadly, this would include family factors (such as structure, social support and parenting), school factors (such as school ethos, support from teachers, enjoyment of school, aspirations), personality measures (such as impulsivity, sensation-seeking and locus of control), peer and friendship factors and other risk behaviours (such as smoking, alcohol, drugs). The availability of such measures at different points throughout the life course is hugely valuable for establishing temporality, in addition to identifying whether the effect of such influences differ according to the age at which the child was exposed to them.

Project outline:

Impact of TTC12-ANKK1-DRD2 gene-cluster on smoking behaviour, impulsivity, and reward in adolescence.

Adolescence is a critical period for smoking initiation (1). Approximately 80-90% of adult-smokers initiate use before age 18 years (2). Early initiation of tobacco use is associated with increased risk of developing long lasting nicotine and other drug dependencies(3). Genetic factors are known to influence risk of initiating and persisting in smoking(4). However, the specific genes mediating this vulnerability are largely unknown. In addition, most of the genetic studies performed so far on smoking have focused on adult and little is known about genetic factors moderating vulnerability to smoking in adolescence (5). Finally little is known about the neurobiological mechanismsunderlying nicotine addiction in youth.

Maternal substance use during pregnancy is highly correlated with serious negative neurodevelopmental fetal and childhood/adolescent outcomes. Research has established maternal tobacco and alcohol use as having the most profound effect on children and adolescents, although the use of other substances such as cannabis (Fried, 2002) and cocaine has also been studied (Huizink & Mulder, 2006). Up to this point, the focus of research on maternal smoking and drinking during pregnancy has primarily been on externalizing and conduct/behavioral outcomes in the child (Fried et al., 1997; Streissguth et al., 1984; Millberger et al., 1996; Fergusson et al., 1993; Knopik, 2009; Knopik et al., 2005; Wakschlag et al., 1997; Weitzman et al., 1992; Rantakallio et al., 1992), including ADHD (Huizink & Mulder, 2006), oppositional defiant disorder, conduct disorder (Fergusson et al., 1998), and later substance abuse. However, findings from recent research have suggested a potential relationship between prenatal nicotine and alcohol exposure and internalizing problems caused by disruptions in neurobehavioral and cognitive development, potentially resulting in depression, generalized anxiety, impaired memory function, lower scores on arithmetic tasks (Batstra et al., 2003) deficits in verbal learning (Richardson et al., 2002; Cornelius et al., 2001), auditory, and visual-perceptual processing (Fried, 2002), as well as IQ decrements (Streissguth et al., 1998, 1990) and slower information processing speeds (Huizink & Mulder, 2006). Evidence also indicates that these adverse effects may persist into early and late adolescence (Wakschlag et al., 1997), resulting in subsequent impaired executive functioning (Huizink & Mulder, 2006), increased incidence of substance abuse (particularly smoking) (Ernst et al., 2001) or a predisposition to early onset of smoking in offspring (after intrauterine exposure to heavy smoking) (Huizink & Mulder, 2006; Ernst et al., 2001; Kandler et al., 1994; Cornelius et al., 2000; Buka et al., 2003). It is still unclear, however, whether these results provide sufficient evidence to suggest a cause-and-effect relationship or if the association is confounded by variables such as social background, child-rearing practices, maternal and family characteristics (Fergusson et al., 1998), or genetic factors.

"Traffic-related air pollution as a risk factor for the development of childhood asthma"

This project, funded by AllerGen NCE, will bring together data on air pollution, and genetic and disease information from existing studies done in Canada, England, The Netherlands, Germany, and Sweden. (See listing below) Data from these studies (all of which have measured the development of asthma in children from birth) will be analysed together to address the question: Does exposure to traffic-related air pollution (TRAP) increase the risk of incident asthma in children? Data from these existing studies will provide important information that can be useful to develop preventive strategies related to childhood asthma and potentially to develop treatment approaches focused on a particularly vulnerable population, children with genetic profiles conferring particular risk.

Project outline:

Research Questions

1) What are the long-term physical and mental health effects ofmedication treatment for ADHD?

2)Does medication treatment for ADHDhave an impact onsocial, behavioural and educational functioning in adolescence?

3) For children meeting criteria for ADHD at age 7 years, which factors best predict persistence of ADHD at age 15 years?

4) What are the prevalence and correlates of apparent 'new' or late onset presentations of ADHD after the age of 7 years?