Our group at the Singapore Institute for Clinical Sciences and the University of Manchester have received funding from the Wellcome Trust to investigate the machine learning modelling techniques for high-dimensional measurements from population studies. Recently, high-throughput 'omics technologies have allowed us to collect large numbers of molecular measurements from a single sample. Analysing these large datasets poses a problem, because existing techniques do not allow us to analyse all measurements jointly as outcomes. We will develop a statistical technique for jointly analysing large numbers of molecular measurements, using recent advances in statistical inference, as well as applying prior knowledge to reduce the number of relationships between measurements that needs to be explored. Additionally, we will develop a method for designing longitudinal studies to gain optimal information about these high-dimensional outcomes. The benefits of our approach will be two-fold: First, in allowing us to gain additional information about the relationships between longitudinal measurements, and second in improving the design of future studies, which will lead to time and cost savings.

We and others have identified numerous factors that influence mental health in young adults and early adulthood. These include obstetric complications, adversity, cannabis use, population density, immigrant status (PMID: 29352556, 12091183, 31563981) and most recently exposure to COVID (PMID: 35987197). Additionally, in the ALSPAC cohort, we recently identified that suicidal ideation at age 17 is associated with 7-fold increased odds of psychotic disorder, and with depressive disorder and generalised anxiety disorder at age 24. Indeed, over 40% of those with psychotic disorder in early adulthood had experienced this symptom in late adolescence (Mongan et al, 2022, unpublished).

We and others have also identified evidence for inflammatory marker dysregulation both preceding and in association with psychiatric disorders including psychosis, depression and recently, long COVID (PMID: 36085284 , 35472304). Dysregulation of acute inflammatory markers [such as Interleukin (IL)-6 and C-Reactive Protein] and complement proteins has been reported prior to and in association with these outcomes (PMID: 25133871, PMID: 32857162). Using ALSPAC data we identified for the first time the cross-sectional association of suPAR (soluble urokinase plasminogen activator receptor- an established marker of chronic inflammation) and IL-6 (an established marker of primarily acute inflammation) with psychotic disorder in young adults (Mongan et al 2022, under review). We previously found an inverse association between the antiinflammatory marker n-3 fatty acid docosahexaenoic acid (DHA) at age 17 and psychotic disorder at age 24 in the ALSPAC cohort (PMID 34059620). Using ALSPAC data we have also shown for the first time that elevated suPAR is associated cross-sectionally with cannabis exposure, itself a marker for mental illhealth (Power et al 2022 (under review)). We now seek to expand our knowledge to mental health outcomes in early adulthood (at age 24 and also at age 30 (when available)). We wish to examine not only associations between risk factors and inflammation in relation to clinical thresholds of psychiatric disorders, but also in relation to specific symptoms and symptom severity.

We hypothesise that exposure to environmental risk factors and early-life adversity is associated with chronic inflammatory dysregulation, in line with recent evidence (PMID: 31682707, 26033244, 34990745). We also hypothesise that inflammatory dysregulation will be more common in those who go on to report mental disorders in adulthood (PMID: 25133871, PMID: 32857162). Crucially, we hypothesise that inflammatory dysregulation mediates the relationship between environmental exposures and adult mental disorders, which could provide evidence for a biological mechanism by which environmental exposures influences the risk of adult mental disorder. Alternatively, environmental exposures and biological risk factors may operate independently, but have a cumulative effect on the risk of mental disorders.

We propose to investigate this using a combination of existing ALSPAC data, and new data derived from assays undertaken by us which assessed levels of the robust marker of chronic inflammation, suPAR, in age 24 plasma samples donated by ALSPAC participants.

Only data from B4168 will be used.

We aim to explore if parental smoking causes offspring to smoke more. Estimating the effect of social transmissibility of traits is methodologically difficult because people relatives generally share similar genetics and environmental exposures. This can make issues around confounding difficult to control for.
Children inherit a random half of each parent’s genetic liability. Mendelian randomisation (MR) is a study design in epidemiology which leverages this to analogise with a randomised controlled trial (RCT). In an RCT some people are randomised to an exposure, and then followed up over a period of time. In an MR study people are randomised to a genetic variant which increases their liability to an exposure at conception, and then recruited into a study some time after birth.
Although understanding the effects of second-hand smoking has important public health implications, it would be unethical to randomise someone to exposure to second hand smoking because of the definite harm on the individual who would have to smoke, and the probable harm on the person exposure to second-hand smoking.
Here we aim to explore an extension of MR to explore the social transmissibility of traits, and specifically whether parental smoking influences their children to smoke. Although we only inherit half of each of our parent’s genetic variants, the other half can still influence us via our parents. For example, if a parent has a variant which causes them to smoke, this variant in effect exposes their child to second-hand smoking. Thus, randomly not inheriting a smoking variant, is analogous to being randomised to second-hand smoking. Our research here will build on theoretical and simulation research already conducted to demonstrating the theoretical utility of this approach.

Cancer is a leading cause of death, responsible for one in six deaths worldwide. Of these, breast is most common cancer and the 5th leading cause of death among all cancers. Breast cancer can be categorised as a hormone-sensitive cancer, along with prostate, endometrial and ovarian cancer, meaning the development of these cancers is driven by hormones. Menopause marks the end of a woman’s reproductive life and is characterised by changes in a woman’s biology including in the levels of sex hormones circulating their body. The levels of these hormones have been implicated in the risk of these hormone-sensitive cancers. The amount and distribution of fat (adiposity) are also known to change across the menopausal transition and in addition are known major risk factors for cancer. What is not currently known is how the changes in hormones and fat, specifically due to the menopause, affect the risk of cancer. Both hormone levels and fat quantity and distribution are known to be regulated, to some extent, by genetics. This project aims to use genetics of changes in hormone levels and fat quantity and distribution to determine their causal role in cancer risk, both independently and through their interaction. This research will help to identify the causal role of risk factors for cancer at a specific time-point which could then be targeted to prevent cancer.

This project aims to improve our understanding of which factors are predictive of mercury levels in pregnant women. Many factors have been reported in isolation, but little work has previously been done to compare them and identify their relative importance. Fish consumption is likely to be a primary source, but there may be a role of social factors (eg job) and genetic makeup.

We and others have identified numerous factors that influence mental health in young adults and early adulthood. These include obstetric complications, adversity, cannabis use, population density, immigrant status (PMID: 29352556, 12091183, 31563981) and most recently exposure to COVID (PMID: 35987197). Additionally, in the ALSPAC cohort, we recently identified that suicidal ideation at age 17 is associated with 7-fold increased odds of psychotic disorder, and with depressive disorder and generalised anxiety disorder at age 24. Indeed, over 40% of those with psychotic disorder in early adulthood had experienced this symptom in late adolescence (Mongan et al, 2022, unpublished).

We and others have also identified evidence for inflammatory marker dysregulation both preceding and in association with psychiatric disorders including psychosis, depression and recently, long COVID (PMID: 36085284 , 35472304). Dysregulation of acute inflammatory markers [such as Interleukin (IL)-6 and C-Reactive Protein] and complement proteins has been reported prior to and in association with these outcomes (PMID: 25133871, PMID: 32857162). Using ALSPAC data we identified for the first time the cross-sectional association of suPAR (soluble urokinase plasminogen activator receptor- an established marker of chronic inflammation) and IL-6 (an established marker of primarily acute inflammation) with psychotic disorder in young adults (Mongan et al 2022, under review). We previously found an inverse association between the anti-inflammatory marker n-3 fatty acid docosahexaenoic acid (DHA) at age 17 and psychotic disorder at age 24 in the ALSPAC cohort (PMID 34059620). Using ALSPAC data we have also shown for the first time that elevated suPAR is associated cross-sectionally with cannabis exposure, itself a marker for mental ill-health (Power et al 2022 (under review)).

We now seek to expand our knowledge to mental health outcomes in early adulthood (at age 24 and also at age 30 (when available)). We wish to examine not only associations between risk factors and inflammation in relation to clinical thresholds of psychiatric disorders, but also in relation to specific symptoms and symptom severity.

We hypothesise that exposure to environmental risk factors and early-life adversity is associated with chronic inflammatory dysregulation, in line with recent evidence (PMID: 31682707, 26033244, 34990745). We also hypothesise that inflammatory dysregulation will be more common in those who go on to report mental disorders in adulthood (PMID: 25133871, PMID: 32857162). Crucially, we hypothesise that inflammatory dysregulation mediates the relationship between environmental exposures and adult mental disorders, which could provide evidence for a biological mechanism by which environmental exposures influences the risk of adult mental disorder. Alternatively, environmental exposures and biological risk factors may operate independently, but have a cumulative effect on the risk of mental disorders.

We propose to investigate this using a combination of existing ALSPAC data, and new data derived from assays undertaken by us which assessed levels of the robust marker of chronic inflammation, suPAR, in age 24 plasma samples donated by ALSPAC participants.

Daily movement behaviours can be broadly classified into four categories, dependent largely on intensity, namely sleep, sedentary behaviour, light physical activity and moderate-to-vigorous physical activity. The government recommend children and young people partake in an average of 60 minutes of moderate to vigorous physical activity every day, in order to maintain good health. Whilst current research supports the need to improve physical activity engagement in childhood, interventions thus far have shown little to no significant improvement. Furthermore, whilst it is suggested that children with asthma follow the same guidelines in regard to engagement in physical activities, but no guidance is given as to the type of exercise, or whether different intensities are more suitable for different clinical groups.

The aim of this project is to characterise the composition, and level of, physical activity behaviours throughout the day, with specific interest in exploring whether those diagnosed with asthma show similar patterns to those without asthma. The data set will be combined with data from the Millennium Cohort Study and the Commando Joes/X4A Trial, and analysis will consider age, sex and asthma status.

Problematic menstrual symptoms, such as pain, heavy bleeding, and irregular cycles, impact a high proportion of women and people who menstruate and are associated with multiple adverse physical and mental health outcomes, as well as reduced attendance and productivity at school/work. Despite this, little research has sought to identify the causes and risk factors associated with such menstrual symptoms. Socioeconomic disadvantage is one factor that has been associated with worse menstrual symptoms; however, the current evidence is mixed and unable to understand causality. It is possible that socioeconomic position (SEP) causally impacts menstrual symptoms due to early life stressors and associated lifestyle factors adversely impacting the development of the brain, the nervous system, and hormone production systems. Additionally, menstrual symptoms could negatively impact SEP through their impacts on school and work thus restricting the ability of women suffering with such symptoms to reach their fully academic and career potential. Therefore, this project aims to understand the causal, bidirectional relationship between SEP and menstrual symptoms by combining observational and genetic methods in multiple UK-based cohorts. Robust evidence that SEP and menstrual symptoms are causally related may support the need for additional support or treatment for individuals from disadvantaged backgrounds and/or provide rationale for improving school and work environments to enable women to better manage problematic menstrual symptoms.

There appears to be a sensitive period from ages 7 to 11 whereby conduct problems and head injuries promote one another. Individually, both have been associated with poor executive functioning such as disinhibition, poor working memory, and deficits in emotion recognition. However, what is not yet known is how this bidirectional association between ages 7 to 11 may affect executive function in adulthood. To address this gap in the literature, we plan to estimate how executive function at age 24 may be predicted by childhood head injuries and conduct problems from ages 7 to 11.