Proposal summaries
B3367 - Maternal nutritional status in pregnancy and offspring Autism Spectrum Disorder a social perspective - 21/09/2019
What a mother eats during pregnancy has previously been established to programme their infants health in later live, even tracking through to adulthood. However, despite compelling evidence to support foetal programming of physical health, neurodevelopment has remained understudied. Recently, autism has gained increased attention within this field of research. Although largely caused by genetic factors, it is now estimated that approximately 20% of the risk of developing autism relates to environmental factors, of which nutrition is a main focus. Preliminary evidence indicates that the risk of developing autism may be reduced through consumption of a nutritionally adequate diet or nutritional supplements. Furthermore, the presence of social patterning is unknown. Health inequities are observed in numerous health outcomes as low socioeconomic position groups experience more adversity including poorer diet quality. Thus, it is possible that, where an association existed between maternal diet and their childâs risk of autism, it may also be socially patterned.
Two datasets will be compared for this project, ALSPAC and a second dataset called the Norwegian Mother and Child Cohort Study. The data from each cohort will be analysed in the same way.
This project will apply statistical methods that estimate if there is an association between maternal nutritional intake during pregnancy and the childâs risk of developing autism. The project will apply âcausal methodsâ which are statistical methods which better estimate causal relationships from observation data when compared to conventional methods with measure âassociationâ. A key way in which this occurs is through adjusting for confounders which vary over time, such as early infant nutrition. Both pregnancy and early infant nutrition such as breastfeeding, complementary feeding and nutrition in the early years have been associated with neurodevelopment and so the effects measured in pregnancy may be further altered by nutrition consumed in childhood. Lastly, results will be stratified by socioeconomic position to more clearly assess the presence and the degree of social patterning in this relationship.
B3366 - Non-alcoholic fatty liver disease and cardiac function in young adults - 06/09/2019
Although there are many common risk factors between non-alcoholic fatty liver disease (NAFLD) and cardiovascular disease, e.g. obesity, NAFLD may be associated with cardiovascular disease, independently from other established cardiovascular risk factors.
B3365 - Specifying Risk for Driven Exercise in Eating Disorders - 02/09/2019
Physical activity can be rewarding and lead to anxiety reduction, and physical activity levels in the general population are heritable, suggesting individual differences in the degree to which physical activity is reinforcing via biobehavioral pathways. Increases in physical activity are often associated with positive mental health outcomes, though more is not always better. For example, up to 40% of individuals with bulimia nervosa and up to 60% of those with anorexia nervosa present with driven exercise (exercising in a driven manner to control weight and shape). Driven exercise is a serious, concerning, and understudied eating disorder symptom. Existing research points towards compulsivity and general propensity to exercise as potential risk factors for driven exercise. At present, our understanding of the developmental etiology of driven exercise is poor, and the extent to which this symptom relates to patterns of PA during development is unknown.
We will examine risk for driven exercise in the ALSPAC cohort by identifying trajectories of physical activity during development and identifying whether these trajectories associate with driven exercise and eating disorder diagnoses in adolescence and emerging adulthood. We will then investigate the developmental timing and predictive strength of driven exercise in relation to eating disorder onset and maintenance. Finally, we will evaluate the degree to which genetic risk profiles inform risk for driven exercise.
B3363 - Large-Scale Evaluation of the Effect of Rare Genetic Variants on Psychiatric Symptoms and Cognitive Ability - 02/09/2019
Rare copy number variants (CNVs) are strongly associated with neuropsychiatric disorders, suggesting that they might serve as a magnifying glass to study general mechanisms of psychopathology as otherwise subtle perturbations to neuropsychiatric functions may be more clearly discerned through the major âhitâ of the CNV. However, our understanding of the impact of CNVs on psychiatric symptomatology, RDoC domains and neurocognitive ability (termed âdimensional neuropsychiatric phenotypesâ) is limited in at least three ways. First, the effects sizes of the vast majority of CNVs on neuropsychiatric phenotypes remain poorly understood and their rarity will likely to prevent individual association studies. Prior studies concentrated on the most recurrent CNVs, leaving more than 90% of these variants undocumented. Second, for CNVs frequent enough to be studied individually, the full spectrum of phenotypic variation is unknown because ascertainment has been performed through neurodevelopmental and specialty clinics, which presumably represent the severe end of the phenotypic spectrum. Only a few studies have been conducted in unselected populations. Finally, many CNVs seem to impact the same neuropsychiatric domains, suggesting a poly/omnigenic model for psychiatric symptomatology, RDoC domains and neurocognitive ability. Based on this hypothesis, our previous work has shown that genetic scores and functional annotations can accurately predict the effect of any CNV on IQ but these approaches have not yet been extended beyond IQ to other dimensional neuropsychiatric phenotypes. We will fill these knowledge gaps with a novel, multidisciplinary, collaborative project that leverages existing archival data (n=255,303) to estimate and predict the effect sizes of CNVs (duplications and deletions) on dimensional neuropsychiatric phenotypes. Our aims include 1) phenotypic harmonization; 2) characterizing previously identified risk CNVs for mental illness in a large in general population cohorts and in samples ascertained for mental illnesses; 3) examine the contribution of common variants to variable expressivity of rare CNVs via polygenic risk scores (PRS) in the domains of mood, psychosis, developmental disability, and general cognitive ability; and 4) develop novel models to explain the effect size of any rare CNVs on dimensional neuropsychiatric phenotypes. Finally, we will develop tools for data sharing.
B3364 - Do womens expectations impact their birth experience and health outcomes - 03/09/2019
Expectations of birth are thought to have important effects on a womanâs experience of birth, and her satisfaction with her care. For her to have a positive experience, her priority expectations need to be met. If there is a gap between her expectations and the reality of her birth, it may impact on her psychological and physical wellbeing postnatally. This gap has been documented widely in the published literature when speaking to women and staff. However, there has been little opportunity to quantify this gap, or the impact it has.
This study provides a unique opportunity to quantify and explore the expectation-reality gap, by using data collected over two decades. The Avon Longitudinal Survey of Parents and Children has been following women who had children in the South West of England in 1991/2, it has now started to follow their daughters having their babies. This means that inter-generational trends can be explored.
This work will allow us to define the expectation-reality gap and explore what causes it, its effects and suggest ways for reducing it in the future.
B3361 - EpiTIME Solving the time puzzle of epigenetic effects on child mental health - 30/08/2019
EpiTIME aims to shine a light on the newly discovered epigenetic âtime puzzleâ of child mental health. Recently, it has been observed that common mental health problems in children, such as inattention-hyperactivity and impulse-control problems, are most strongly predicted by epigenetic patterns regulating gene expression at birth â a signal that is curiously lost when measuring these same patterns later in childhood. Such a finding points to the existence of an early biologically-sensitive developmental window and may provide us with crucial insights into the nature and origins of mental health outcomes in children. Yet, how these epigenetic timing effects arise, what factors drive them and why they manifest is currently a puzzle. To solve it, this project will combine (i) the application of innovative, multidisciplinary approaches and (ii) the generation of new data within a unique set of European longitudinal cohorts to systematically characterize, locate and explain epigenetic timing effects on child mental health with unprecedented scale and depth. As well as addressing a major knowledge gap and advancing research at the forefront of biological and psychological sciences, EpiTIME has the potential to set in motion a paradigm shift in the way that we conceptualize, understand and approach mental health in children.
B3362 - Correlating whooping cough susceptibility and pertussis vaccine immune responses through HLA diversity - 05/09/2019
Whooping cough is a vaccine-preventable disease that has the potential to cause significant morbidity and mortality in unvaccinated individuals. Despite the success of the vaccine there are recent reports of disease in older adolescents and young children who have been vaccinated and the causes for these failures are unknown but are likely to stem from our poor understanding of exactly which components of the bacteria causing the disease (B. pertussis) should be targeted. The ALSPAC team have recently published a study demonstrating that genetic differences in a key region of the human genome, the human leukocyte antigen (HLA) complex, may be associated with differential susceptibility to whooping cough. We have similarly undertaken a genetic study of African children finding associations across the same region of HLA with differential response to three different parts of the whooping cough vaccine. We would like to use sophisticated genetic techniques to compare our results with those from ALSPAC to determine whether we can show that reponses to one or several vaccine components is related to whooping cough susceptibility. These results will enable us to understand why the vaccine is failing in some groups of individuals and how we can improve the vaccine for multiple populations in the future.
B3360 - Genome-wide meta-analysis of infant developmental milestones and temperament - 21/08/2019
Temperament broadly refers to individual differences in behaviour that are typically measurable in infancy and early childhood. Broad dimensions include domains such as emotionality, negative affect, sociability and surgency. Measures typically capture a large number of individual subscales, as well as more general overall domains.
Fortunately, measurement of infant temperament has been developed over several decades, with considerable number of psychometric studies to support the measures and with an emphasis on capturing reliable individual differences. Commonly used scales include the infant temperament scales by Carey and the InfantâToddler Social and Emotional Assessment by Carter and Briggs-Gowan.
Temperament and developmental milestones reflect early development of personality and behaviour. Infant temperament and milestones predict a variety of later outcomes in childhood.
Twin heritability for temperament domains has tended to be reported as between 30-40%. In general, this research field is characterised by smaller twin studies compared to studies of older ages. Some of the very large developmental twin cohorts of >5000 pairs (e.g. TEDS, CATSS) have either tended to begin main assessments after infancy or have included a small number of items in infancy.
Our study aims to explore the role of genetic variants on infant temperament and developmental milestones using a variety of state of the art statistical genetic methodology.
B3359 - Predicting adult height among children with idiopathic short stature using a polygenic risk score - 19/08/2019
Children with idiopathic short stature (ISS) are defined by height below 2 standard deviations (SD) of the mean for age and sex without any endocrine, metabolic or other disease explaining the short stature. The US Food and Drug Administration approves growth hormone (GH) treatment on children shorter than 2.25 SD of the mean for age and sex with a predicted adult height below the normal range. Given that stature in a population follows a Gaussian distribution, 2.3% of children will always be shorter than 2 SD below the mean for age and sex. However, a proportion of these children defined in childhood as having ISS will eventually achieve a normal adult height or a normal height in their families, even in the absence of expensive GH treatment.
Human height has a highly polygenic nature. It has been estimated that about 80% of variation in height can be attributed to genetics. Polygenic scores have been demonstrated to have an improving ability to identify individuals at significantly high/low predisposition towards complex diseases. Therefore, it has become possible to identify individuals who will lie at the extreme distribution of a trait, such as height.
Therefore, we posit that a polygenic risk score for adult height may be able to effectively predict which children diagnosed as having ISS are likely to achieve a normal adult height where indication of GH treatment would not be necessary.
B3358 - Trajectories of hearing and cognitive function through the lifecourse in the Avon Longitudinal Study of Parents and Children - 30/08/2019
Loss of hearing is common, and tends to increase over the life course, affecting over 10 million people in the UK. Although there have been many studies concerning deafness in childhood, very few have examined the normal course and variation in hearing in a large population of individuals from an early age into mid and older adulthood. This is troubling since even low levels of hearing loss can result in failure to hear speech clearly, and can impact social communication, mental health and employment. Moreover hearing loss is associated with cognitive decline, and has been identified as one of the nine modifiable risk factors in the Lancet Commission on Dementia. It is not known whether hearing loss is a causal factor for cognitive decline, or a non-causal feature associated with ageing and neurodegeneration. Monitoring hearing in a population from early in life, and using novel genetic methods to investigate causality, may be crucial to understanding not just the role of hearing in cognitive function, but also the ageing process in general.
Utilising the Avon Longitudinal Study of Parents and Children (ALSPAC), this project will be the first to study, in depth, the changes in hearing ability over the life-course from early childhood to age 30 and to examine the relationship with cognitive function.
Firstly we will analyse data from approximately 5000 individuals, for whom we have detailed measures of hearing function at ages 7, 9, 11 and 14, and which we will collect again when they are age 30. We will characterise how their hearing has changed from age 7 to age 30, and identify those with who have experienced a drop in hearing ability. We will examine whether changes in hearing are associated with environmental exposures or the presence of particular genes.
Secondly we will analyse their cognitive function, using measures of memory, attention and processing speed collected at age 8, 10, 13, 15 and 24, and which we will collect again at age 30. We will compare trajectories in hearing ability with trajectories of cognitive function from age 7 to age 30, and test whether those who experience a decrease in their hearing ability are more likely to have poorer processing speed, attention and working memory at age 30.
The project is unique in that:
⢠it will provide observational information on the natural history and genetic influences on hearing over the first 30 years of life
⢠it will be the first to assess age-related relationships between changes in hearing and features of cognition through childhood into early adulthood
The information collected will be valuable for studies of further ageing of this population as well as identifying possible mechanisms linking hearing and cognition.
B3357 - Pubertal development and the gender gap in education A study using the ALSPAC birth cohort - 16/08/2019
Educational attainment and achievement has increased significantly among both men and women in industrialised countries over many years. At the same time, an increasing gender gap in education has developed in favour of women (OECD, 2015). The gender gap in education now represents a societal challenge in many industrialised countries. The causes are not known and under-researched, and there has been limited attention to potential policies for decreasing the gap.
Girls enter puberty earlier than boys, and by age 15 to 16 the gender difference in maturity reaches a peak (Mustanski et al, 2004). At this age, adolescents graduate from lower secondary education and in many European countries (including the UK and in the Nordic region), it coincides with important decisions regarding their future. The opportunities available to individuals depend heavily on their grade scores from lower secondary school. An important question to ask is therefore whether and to what extent the âbiological head startâ of girls explains their educational outperformance of boys.
B3352 - Protective factors in the association between exposure to domestic violence in childhood and internalising symptoms - 12/08/2019
Childhood exposure to domestic violence is associated with long-term impairment such as increased risk of mental health illness, aggression, anti-social behaviour and poorer academic attainment, yet some children function well despite this adversity. As part of my doctoral thesis, I propose to use ALSPAC to identify the key factors which protect children and adolescentsâ mental health following exposure to domestic violence. I will also examine whether or not the protective effect of these factors vary by socio-demographic and contextual factors (i.e. age, gender, socio-economic status, and severity/duration of domestic violence).
Resilience is the ability to bounce back and successfully adapt to challenging circumstances. It is not a personality trait and is amenable to change. Individual, family, and community protective factors that promote resilience in the face of childhood adversity have been identified within the literature . Studies that have explored protective factors within the contexts of domestic violence exposure and mental health outcomes have tended to explore single factors on their own, have not considered the complexity of the issue or contribution of other adversities, and have not necessarily considered the severity of the violence witnessed. Additionally, the vast majority of all research in the field has been conducted in the USA where attitudes towards violence and outcomes for both parents and children may differ to the UK. Therefore, more evidence is needed to identify how these factors protect against mental disorders in children who have been exposed to domestic violence. This research will provide an understanding of protective factors and their effects in different contexts will provide better information for the development of targeted interventions aimed at improving the mental health of children exposed to domestic violence. Furthermore, identifying whether these protective factors only buffer against poor mental health under certain conditions will help us determine whether such interventions should be tailored to children and young people based on their characteristics and trauma exposure.
Alongside this, we would like to identify whether these factors (if any), protecting children and adolescents from internalising symptoms, are specific to the contexts of domestic violence exposure and mental health or whether they may be protective against other behavioural problems and in the context of exposure to other adverse childhood experiences such as parental alcohol/substance abuse, parental mental health problems, direct child abuse and neglect. This will also inform the potential scope of future interventions.
B3353 - Understanding determinants of Telomere length in early life and its effect on Cardiovascular risk throughout the life course - 16/08/2019
Telomere length (TL) is a DNA marker of biological age in humans. Shorter TL (signifying older biological age) associates with higher risk of age-related disease such as coronary artery disease. Whilst TL is strongly heritable, it is also associated with lifestyle and environmental factors, such as diet, exercise and smoking in adults. However, recent studies have proposed that TL is "set" in early life and that environmental/lifestyle exposures may have more effect on TL during childhood. For example, adult smokers have, on average, shorter TL than non-smokers but smoking does not increase TL loss over time in adulthood. Therefore, the relationship between smoking and TL is more complex and is likely established at an early age. One possible explanation is that childhood exposure to smoke from parental smoking both shortens TL in the child and increases the likelihood of them smoking in later life. These same relationships may also be seen for other traits such as diet and exercise, traits that in adults are often influenced by childhood experience. Currently there are few studies that are able to address such questions. By measuring TL in children and young adults at large scale (minimum 2,900 per age group, 7, 17 and 24 years) we can more accurately explore the relationships between lifestyle and environment in childhood with TL. This will provide important information about how TL is influenced in early life and how early life exposures influence disease risk in adulthood through a biological ageing mechanism.
B3354 - Understanding the evolution of joint hypermobility pain and associated symptoms a novel multigenerational longitudinal cohort - 22/08/2019
B3355 - Anorexia and subsequent smoking cause or correlation - 09/08/2019
Smoking prevalence is reported to be increased in populations with anorexia nervosa (AN), and engagement in smoking in AN is suggested to result from attempts to control weight. However, both smoking and AN have also been linked with earlier anxiety; thus anxiety may explain the association between smoking and AN. This study will assess the prospective association between AN and subsequent smoking, across three longitudinal waves of data. Models will be adjusted for childhood worry (a symptom central to anxiety disorders), to determine whether AN explains smoking beyond the predictive effects of anxiety pathology.
B3356 - GWAS of breast density - 09/08/2019
Increased breast density is strongly associated with increased breast cancer risk, however, the underlying biology is unclear.
We aim to identify genetic variation associated with breast density to help explain this association.
B3348 - Testing a model of whether early non-specific symptoms independently predict unhealthy lifestyle behaviours and psychosis - 06/08/2019
Psychosis refers to the experience of hallucinations and/or delusions. Psychotic experiences range from short-lived symptoms that are not fully believed through to persistent severe symptoms that characterise psychotic illnesses as schizophrenia. It is well established that enduring psychotic illnesses such as schizophrenia have worse physical health than the general population, and are at an increased risk of developing long-term physical health conditions such as diabetes and heart disease. This may be due to unhealthy lifestyle behaviours, such as a lack of physical activity and smoking. Such risk factors begin early in first-episode psychosis and even before the first episode. However, the causal relationship between poor physical health and psychosis is not fully understood. We think poor physical health and psychosis may occur independently of each other, but are both related to earlier more common mental health difficulties, such as depression and anxiety.
Not everyone who experiences mental health difficulties will go on to develop psychosis or physical health problems. For those who do, recovery is possible. Medication and supportive relationships are known to help with recovery, however less is known about the role of romantic relationships specifically.
This research project aims to explore how these factors interact over time within the ALSPAC dataset.
B3349 - Simulated ALSPAC data as a resource for longitudinal research and teaching - 05/08/2019
The Avon Longitudinal Study of Parents and Children (ALSPAC) is a unique resource with a wealth of rich longitudinal data. Furthermore, ALSPAC is one of the few longitudinal cohorts with repeated assessments of self report psychiatric traits, along with a host of early exposures and later outcomes. As such, ALSPAC is a vital tool for exploring the longitudinal nature of psychiatric traits, their antecedents and later consequences.
Examining the nature of psychiatric disorders such as depression is complex and often requires advanced statistical methods to untangle complex associations and underlying mechanisms. Currently, there are few open access datasets with enough detail available for researchers to learn these complex statistical methods. As such, many researchers are forced to use datasets that do not capture the complexity of traits such as depression, and this could hinder the ability to make further progress in uncovering diseases like depression.
Given the sensitivity of the ALSPAC study, it is not appropriate to release full versions of the data. However, it is possible to simulate parts of the ALSPAC data to give the same properties, without the risk of disclosure and identification of participants.
Simulating ALSPAC data that matches the original properties of the data would provide an excellent resource for teaching purposes as well as providing an introduction to researchers wanting to use the original ALSPAC study.
B3351 - Broad Antisocial Behavior Consortium BroadABC - meta-analysis of antisocial phenotypes - 05/08/2019
The BroadABC has been created to combine the results of multiple genome-wide association studies of broad antisocial behavior (measured by symptom counts of antisocial personality disorder, ratings of aggression, conduct problems, delinquency, psychopathic personality etc) in meta-analyses in order to increase the probability of detection of genetic variants associated with individual differences in liability to antisocial behaviors. For phase 2 our aim is to include at least ~150,000 individuals.
B3350 - The Heart-Brain Connection in ALSPAC30 Cardioaggression or Neuroselection - 08/08/2019
In our ageing population the burden of both heart failure (HF) and dementia is increasing. To date there is no proven preventative or curative treatment for cognitive decline and the associated social and economic cost is huge. Both conditions share common risk factors, yet there is growing evidence of a direct relationship between the function of the heart and the brain. But we still don't know if poor cognition is a consequence or a cause of poorer cardiac function. Several aspects of this association require thorough investigation. We propose that there is a bidirectional association between cognition and cardiovascular disease (CVD) and by applying sophisticated techniques to assess cardiac and brain structure and function, this investigation will significantly advance our understanding of the causal mechanisms underlying both cardiac and cognitive decline.