GWAS have been enormously successful in uncovering novel genetic variants associated with a range of complex human diseases, but the majority have used cross-sectional data and relatively simplistic statistical tests. Longitudinal studies are advantageous for investigating genetic associations as they: 1) facilitate the detection of genetic variants that influence change in a trait over time; and 2) allow the detection of genes that are associated with the age of onset of a trait. Improving analytic techniques for conducting longitudinal GWAS offers the opportunity to advance our understanding of the aetiology of health and disease.

To test statistical methods for modelling trajectories, we have selected two growth traits: height and body mass index (BMI). These traits were selected as most cohorts within EGG (The Early Growth Genetic consortium) have data available for analysis, therefore maximizing our sample size. Height was chosen as its trajectory has a relatively simple (linear) shape, whereas BMI is more complex but is of interest to a large number of groups within the consortium. Ideally, the methods we define will be transferable to other traits of interest across the consortium.

Childhood obesity is associated with poor mental and physical health and is one of the strongest predictors of adult obesity. BMI, the most commonly used quantitative measure of adiposity, follows a well-characterised trajectory throughout childhood: a rapid increase soon after birth until approximately 9 months, the adiposity peak, followed by a gradual decline until around 4–6 years of age, the adiposity rebound, followed by an increase again until the end of puberty. Although the community has identified a large number of genetic variants associated with adult BMI, relatively little is known about the genetic determinants of BMI throughout infancy and childhood, or the rate of growth across early life.

Distinct height growth patterns, particularly during puberty, have also been linked with adverse health outcomes such as poor cardio-metabolic health. Analysis in the EGG consortium using SITAR growth curve analysis has shown that velocity of height growth throughout puberty is genetically correlated with adult health, including glycemic traits, metabolites, bone density measures, lung function and lung cancer. By investigating height growth across childhood, rather than just puberty, we will be able to investigate whether these genetic correlations are more generally related to early life growth.

Typical analysis of genetic data to find a gene with a risk for a disease is designed to look for single gene/allele relationships. While multiple relationships can be found they assume independence. We have developed novel methods that assess allele-allele interactions. Normally in biology genes interact and abnormal changes in gene expression or functions in two or more members of an interacting set of genes may lead to disease. OUr novel methods can build netwoks of allels that combine to increase risk of preeclampsia and potentially other diseases of pregnancy.

Nonverbal Learning Disability (NVLD) is a disorder characterized by deficits in visual-spatial, but not verbal reasoning, as well as impairment in two of four areas: math calculation, visual executive functioning, fine motor skills, or social skills. Prior findings established the population prevalence of NVLD to be 3% - 4% in children and adolescents.(1) We aim to establish the stability of NVLD over time, using children’s test scores at two time points (age 7 to 9 and again at age 11 to 15). This project also identifies indicators and risk factors of NVLD, such as child temperament and environmental exposures, and studies NVLD's long term outcomes.

References

1. Margolis AE, Broitman J, Davis JM, Alexander LM, Hamilton A, Liao Z, Banker S, Thomas L, Ramphal B, Salum GA, Merikangas KR, Goldsmith J, Paus T, Keyes K, Milham MP. Estimated Prevalence of Nonverbal Learning Disability Among North American Children and Adolescents. JAMA Network Open. 2020;3(4):e202551-e202551.

In this study I am going to look at how different lifestyle factors, like foods eaten, timing or frequency of eating, physical activity, sedentary behaviours and their timing or location, combine together to create an overall behavioural pattern score that indicates whether adolescents have good health. It has been previously found that a combination of factors is more important for health compared with single factors alone. I also plan to use a new, reproducible laboratory technique, known as metabolomics, to record over 220 measures of blood that indicate a range of metabolic processes. This will help to find out in much more detail than ever before how behaviour leads to better cardiovascular health via metabolic pathways. When it is known more about the pathway that leads from lifestyle to disease it will be easier to predict who will stay healthy and who will not from their behaviours.

I will be using information from diet diaries, activity monitors and blood samples that have already been provided by Children of the 90s participants when they were teenagers.

Since the coronavirus disease (COVID-19) outbreak in December 2019, changes to societies have been observed globally due to rising infection rates and fatalities. Despite economic risk, public health measures have been enforced, including lockdown(s), social distancing, self-isolation, shielding, and the closure to schools and businesses. We already understand that participants from the ALSPAC cohort have shown higher rates of anxiety. Therefore, this study hopes to determine what influence anxiety has had on obesogenic dietary patterns (energy-dense, high-fat and low-fibre diets).

Selection bias is pervasive in human studies and even more so for epigenome-wide association studies (EWAS) due to the relatively high costs of measuring DNA methylation. Although this bias is often acknowledged, its impact on findings has not been evaluated. We aim to perform such an evaluation for EWAS performed in ALSPAC. We will characterise selection by comparing the characteristics of ALSPAC participants measured at baseline with and without DNA methylation profiles and then estimate the resulting bias using simulations informed by the data.

Most measures in epidemiology are subject to error; despite this, the impacts of measurement error are often overlooked. Measurement error in an outcome variable will generally result in a loss of precision whereas measurement error in an exposure variable (or other covariate) will generally lead to bias in regression coefficients.
In this project we will quantify measurement error in systolic blood pressure, weight and height using repeated measurements made on a subsample of individuals attending ALSPAC clinics. We will then assess the impact of measurement error in these variables on regression coefficients and their standard errors.

Paediatric cerebral visual impairments (CVIs) are vision problems caused by damage to the brain rather than the eyes. Injuries to the newborn brain are by far the most common causes, including injuries associated with prematurity, hypoxia-ischemia and/or infections in the perinatal period. Many of these children have multiple neurodevelopmental impairments affecting movement, cognition and/or behaviour, and require extra educational support. There is concern that in some of these children CVIs go unnoticed or mistaken for other problems. Recent work by our group has shown that CVIs might be more prevalent than previously appreciated in the population, highlighting the need for a panel of biomarkers/predictors which can promote earlier and more accurate identification of high-risk children.
Genetic variants may affect how the newborn brain responds to perinatal stresses capable of causing brain injury. Compared to other neurodevelopmental disorders, research into the genetic contributions to CVIs is lacking, although a few genetic associations have been reported (1, 2).
This study aims at evaluating whether candidate genetic variants in the main candidate pathways involved in perinatal brain injuries (glutamate signalling and inflammation) are associated with CVIs. This hypothesis has derived partly from work carried out by two of the applicants (CW and KL) in the DRIFT study(3), in which children were assessed 10-years after a new intervention to treat brain bleeds. A high proportion of these children had CVIs and the number of CVIs correlated with the degree of structural damage to the brain that the children sustained in infancy. The genetic variants we will investigate have been identified both from the wider literature and from our own previous work (4). This includes preliminary findings within the Bristol Neonatal Gene Study, implicating candidate genetic variants in the glutamate signalling and inflammation pathways in childhood motor and cognitive outcomes. For some of these variants, us and others have explicitly shown functional effects on glutamatergic and inflammatory regulation in in vivo models relevant to brain injuries.
Associations between candidate genetic variants and vision outcomes in children from the general population will be explored within ALSPAC. By combining research from the Bristol Neonatal Gene Study with the rich database of visual function data within ALSPAC, we will be able to consider of a range of potentially important perinatal factors (e.g. prematurity, birth complications, evidence of brain injuries during the neonatal stay in the hospital) that may modify the effect of genotype on visual outcomes. We will also examine the associations between the same SNPs with (a) cognitive and (b) motor findings. From the existing literature we expect there is to be associations and we will compare the effect sizes (ES) in analyses of associations between the vision data and the SNPs of interest, with those obtained when using the motor and cognitive data. These results will be used to onform designs for future powered studies.
Additionally, we will collect new prospective data for children at risk for CVIs, enrolled by the PI at the Bristol Eye Hospital. This new clinical data will include additional more in-depth vision assessments compared to those routinely offered in the standard clinic, which will be evaluated as novel biomarkers for CVIs. Genetic profiles of these children with clinical diagnoses of CVIs will be compared to those from ALSPAC representing the general population.
This work is a collaboration between research groups in Bristol and is complementary to a proposal already approved, submitted in 2019 by SR. Some of the work proposed here is already approved as part of that proposal - however use of the vision data is unique to this proposal

References

1. Bosch DG, Boonstra FN, de Leeuw N, Pfundt R, Nillesen WM, de Ligt J, et al. Novel genetic causes for cerebral visual impairment. European journal of human genetics : EJHG. 2016;24(5):660-5.
2. Bosch DG, Boonstra FN, Reijnders MR, Pfundt R, Cremers FP, de Vries BB. Chromosomal aberrations in cerebral visual impairment. European journal of paediatric neurology : EJPN : official journal of the European Paediatric Neurology Society. 2014;18(6):677-84.
3. Luyt K, Jary SL, Lea CL, Young GJ, Odd DE, Miller HE, et al. Drainage, irrigation and fibrinolytic therapy (DRIFT) for posthaemorrhagic ventricular dilatation: 10-year follow-up of a randomised controlled trial. Arch Dis Child Fetal Neonatal Ed. 2020;105(5):466-73.
4. Pregnolato S, Chakkarapani E, Isles AR, Luyt K. Glutamate Transport and Preterm Brain Injury. Front Physiol. 2019;10:417.
5. Williams C, Northstone K, Sabates R, Feinstein L, Emond A, Dutton GN. Visual perceptual difficulties and under-achievement at school in a large community-based sample of children. PLoS One. 2011;6(3):e14772.

Research has discovered that on any given day in England, 28.5% of young people experience mental health problems, and that as little as one in four of them receive formal support for these problems. There is a lack of knowledge of what happens to those young people not receiving mental health services. For those individuals receiving mental health services, mental health problems have been shown to limit economic, vocational, and social functioning. International studies suggest that 50 to 70% of young people who receive services for their mental health problems continue to experience these problems in adulthood.

The proposed project will learn from cohort data which young people are less likely to receive professional support for their mental health problems and what the characteristics are of those young people not receiving mental health services for their problems. It will also be determined how many young people who do not receive mental health services for their problems continue to experience mental health problems in young adulthood and how resilient these young people were during the COVID pandemic.

The feedback provided by Young People will help direct the course of the project. While researchers from the University of Southampton will share the findings with the academic community, YOUNGMINDS will spread the word about our research and findings to their members and followers, young people, mental health service providers and policy makers using a variety of different media. Output activities will increase awareness and promote discussion among the stakeholder groups and ease the way for more effective mental health support for young people.