|Project Title||Plain Language Summary||Keywords||Applicant||Meeting date|
|Utilising machine learning approaches for comparing the contribution of different types of data for predicting an individual’s risk of ill health||Social surveys are collecting an increasing amount of information about individuals. While we have more information than ever before, we have little understanding of the relative importance of such information to be able to identify which individuals will develop particular health outcomes. Data from Understanding Society will be selected for the following ‘types of data’: personal (e.g. age, sex), socioeconomic (e.g. occupation, education), health (e.g. body mass index, grip strength), biomarkers (e.g. testosterone, cholesterol) and genetic. Machine learning, whereby computers are trained to identify complex patterns in data that are difficult for humans to programme, will then be used to evaluate how well each data type predicts the incidence of ill health among individuals. The main output of the project will be a comparison which data type (or combination of data types) and variables are the best predictors of ill health, allowing policy makers to prioritise their data collection.||Machine learning; predictive model; health||Dr Mark GREEN||4/24/2018|
|University of Liverpool||5/3/2018|
|n/a (named advisers are not accessing data)||5/22/2018|
|Understanding Society Biomedical Data Fellowship|
|An assessment of the total variance explained in HbA1c levels by newly discovered genetic loci||The glycated haemoglobin (HbA1c) test gives your average blood sugar levels over the previous two to three months. It is an important marker of how well individuals regulate sugar levels and can be used to diagnose diabetes. In those already diagnosed with diabetes, the results can indicate whether the measures being taken to control their diabetes are working. Levels of HbA1c are thought to be controlled by a combination of genes and the environment. New research aims to identify genes that play a role in controlling HbA1c and, in this project, we plan to assess how important these identified genes are in explaining HbA1c levels. UKHLS represents a valuable data resource for this work. By exploring the genes that play a role in regulating HbA1c, we can better understand its use (and limitations) as a marker for control of blood sugar levels.||HbA1c, genetics, GWAS||Prof Nicholas TIMPSON||3/5/2018|
|U of Bristol||3/8/2018|
|Laura CORBIN, Eleanor WHEELER, Gaelle MARENNE, Ji CHEN, Ines BARROSO||4/10/2018|
|Genetic and epigenetic associations of iron stores||Iron is an essential element involved in a variety of biological processes. It is tightly regulated to ensure a balance between dietary absorption, transport, storage, and use in order to maintain stable healthy conditions (homeostasis). Blood donors are at an increased risk of iron deficiency due to repeated donation. Understanding the factors affecting an individual’s iron stores may help stop this occurring. Ferritin (a major iron storage protein that reflects the body’s total iron stores) has been measured in Understanding Society and we will use this, with similar data from other studies, to uncover genetic variants (natural differences in DNA between people) associated with iron levels. Furthermore, we will examine regulatory variations in DNA (DNA methylation) of individuals with different body iron stores and “genetic iron scores”. Our aim is to gain new insights into iron homeostasis and how disruptions in the system lead to disease.||iron, ferritin, genetic, risk, blood||Dr Emanuele DI ANGELANTONIO||1/22/2018|
|University of Cambridge||2/2/2018|
|Dr Steven BELL, Dr Adam BUTTERWORTH, Prof Nicole SORANZO, Prof David ROBERTS, Prof Willem OUWEHAND, Prof John DANESH||2/28/2018|
|The National Institute of Health Research (Blood and Transplant)|
|The genetics of favourable adiposity||We would like to access the Understanding Society genetic data and some trait data in order to identify “favourable adiposity” genes. By this we mean versions of genes (alleles) that result in a higher BMI and body fat but lower risk of diseases such as type 2 diabetes, hypertension and heart disease. Current studies have already identified 14 alleles associated with higher body fat percentage but lower risk of these diseases. We know that some of these “favourable adiposity” genes operate by putting more fat in the lower body, resulting in a lower waist to hip ratio (“pear” rather than “apple” shape). We would like to identify more of these alleles through a genome wide association study technique called “multivariate” analysis. This approach identifies alleles associated with higher body fat but a favourable profile of blood-based markers of body fat – for example lower triglycerides and fatty liver proteins but higher good cholesterol (HDL). We will also look to see how alleles that increase body fat link to fertility, as there is a known link between body fat and pregnancy/fertility complications. These alleles will teach us about the body’s ability to store fat in the safest places.||Body fat percentage (%), waist, LFTs, lipids, Blood pressure, adiposity, waist
|Prof Celia LINDGREN||1/22/2018|
|University of Oxford||2/2/2018|
|Hanieh YAGHOOTKAR, Jon MILL, Eilis HANNON, Tim FRAYLING, Sara PULIT||4/6/2018|
|The genetics of favourable adiposity||We would like to access the Understanding Society genetic data and some trait data in order to identify “favourable adiposity” genes. By this we mean versions of genes (alleles) that result in a higher BMI and body fat but lower risk of diseases such as type 2 diabetes, hypertension and heart disease. Current studies have already identified 14 alleles associated with higher body fat percentage but lower risk of these diseases. We know that some of these “favourable adiposity” genes operate by putting more fat in the lower body, resulting in a lower waist to hip ratio (“pear” rather than “apple” shape). We would like to identify more of these alleles through a genome wide association study technique called “multivariate” analysis. This approach identifies alleles associated with higher body fat percentage but a favourable profile of blood-based markers of body fat – for example lower triglycerides and fatty liver proteins but higher good cholesterol (HDL). These alleles will teach us about the body’s ability to store fat in the safest places.||Prof Tim FRAYLING||1/22/2018|
|University of Exeter||2/2/2018|
|Hanieh YAGHOOTKAR, Jon MILL, Eilis HANNON, , Sara PULIT||3/7/2018|
|Unemployment and mental health: A sociogenomic approach||The aim of this project is to investigate the link between underlying genetic factors that are related to mental health and employment and how different social backgrounds and environments (including economic) might influence this relationship. We draw from recent genetic discoveries from large studies who have uncovered multiple genes related to depression, neuroticism and bipolar disorder. Specifically, this project aims to answer the following questions. First, how is a person’s genetics in relation to mental health linked to (un)employment histories. Second, how is the influence of these genetic factors buffered by their social backgrounds and environments, the nature of their job or previous stressful life events. This research works towards a deeper understanding of the complex relationship between genes and our social environment. Better evidence on how social factors moderate genetic influences may broadly benefit our society.||Unemployment, Mental Health, Sociology, polygenic socres, G x E||Prof Melinda MILLS||1/22/2018|
|University of Oxford||2/2/2017|
|Evelina Akimova; Riley TAIJI||3/26/2018|
|ESRC/NCRM (National Centre for Research Methods)|
|Gene-Environment Interplay in the Generation of Health and Education Inequalities||Inequalities in education and health are pervasive, persistent and deeply intertwined. Neonatal health affects education achievements, and education success is associated with longer life expectancies. These inequalities are well documented, and better understanding of the mechanisms underlying them could increase the potential to address them through public policies.
Understanding Society is one of only a few datasets that contains genetic data from people of different ages, under different policy exposures, and follows the same people over time. Our project intends to investigate how genes interact with a dynamic environment (e.g., changing smoking policies, changes in the working family tax credits, temporal variation in exposure to influenza, etc.) to generate inequalities in education and health behaviours.
We will go beyond the old nature versus nurture debate by testing three new hypotheses: 1) adverse shocks in early childhood (e.g., exposure to air pollution, or war) interact with genetic predisposition for smoking, drinking, and educational attainment, 2) public smoking bans differentially affect people according to the genetic predisposition to smoking, and 3) changes in government welfare payments affect people differentially according to genetic predisposition.
We innovate by combining methods from genetics and social science. Building on the discovery of genetic variants that exhibit strong and replicated associations with behavioural outcomes, we will grasp unprecedented opportunities to fill the gap in knowledge about the combined role of genes and environments in causing inequality.
We will examine how Genes and the Environment (GxE) interact to generate inequalities in education and health over the life course. We will go beyond the old nature versus nurture debate by testing two novel hypotheses: (i) children born into advantaged environments are better able to reach their genetically conditioned education potential, and (ii) a privileged environment protects against genetic susceptibility to risky health behaviour.
We innovate by combining methods from genetics and social science. Building on the discovery of genetic variants that exhibit robust associations with behavioural outcomes, we will grasp unprecedented opportunities to fill the gap in knowledge about the combined role of genes and environments in causing inequality. For example, do public smoking policies mitigate the genetic susceptibility to smoking?
|Inequality; Gene; Environment; Interaction; Education; Health; Health Behaviours||Prof Hans van KIPPERSLUIS||10/18/2017|
|Erasmus University Rotterdam||10/24/2017|
|Rita PEREIRA; Niels RIETVELD; Stephanie von HINKE; Pietro BIROLI; Tonu ESKO||10/26/2017|
|NORFACE (New Opportunities for Research Funding Agency Co-operation in Europe – Dial programme|
|Understanding the genetics of neurodevelopmental disorders||We are studying a large group of children with severe intellectual disability, referred from genetics clinics across the UK, in order to find the genes that cause their disorders. We have found evidence that genetic variants (differences in DNA sequence between people) that are common in the healthy population also influence risk of having severe intellectual disability. Some of this common variation is also known to affect educational attainment, i.e. how many years of school someone attends. We want to use the genetic data from people in Understanding Society with information about their educational attainment to compare this to our group of children with intellectual disability. Our eventual goal is to improve our understanding of the biology underlying severe intellectual disability, and to improve our ability to predict the chance that parents with one affected child might have another.||genetics, intellectual disability||Jeffrey BARRETT||9/12/2017|
|The Wellcome Trust Sanger Institute||9/19/2017|
|Wellcome Trust / Dept. of Health|
|Characterising the genetic basis of mental health related symptoms in the general population||While most people do not have psychiatric disorders, it is possible to experience mental health symptoms that are related to some of these disorders, for example feeling low or anxious for a period of time. In the UKHLS study, participants answered a questionnaire about this kind of symptoms. We would like to study the biological background of these symptoms. Specifically, we are trying to find out how much the biological factors causing mental health symptoms overlap with the ones causing psychiatric disorders. The way we approach these biological factors is to compare the genetics between mental health symptoms and psychiatric disorders. If they overlap, it may be possible to learn more about psychiatric disorders in cohort studies that have collected genetic and mental health data, even if they have no direct information on psychiatric diagnoses. Therefore, our study can provide crucial guidance for future research to improve diagnosis and treatment of psychiatric disorders.||biomarkers; gene-environment interactions||Dr Karoline KUCHENBAECKER||9/12/2017|
|University College London||9/19/2017|
|Family background, genetics and educational achievement. A follow-up study.||A number of studies have found big differences in educational achievement between individuals from high and low income backgrounds. Narrowing this achievement gap has now become an issue of great public policy concern. The presumption by many policymakers is that such differences in educational achievement are driven simply by the different quality and quantity of resources (e.g. books, schools, tutoring) that high and low income families can provide. However, an alternative explanation is that at least part of this achievement gap is due to genetic differences.
I will explore this issue using UKHLS data. There are almost 100 genetic variants associated with educational achievement. I will consider whether parts of the DNA code which have been shown to be related to educational achievement differ between high and low income groups. I will then investigate the extent that accounting for these genetic differences can explain variation in educational achievement between individuals from high and low income backgrounds.
|Educational achievement, socio-economic gaps||Prof John JERRIM||9/12/2017|
|The Genetic Basis of Political Behaviours and Orientations||The aim of this project is to develop numerical scores for combinations of genes that are associated with
educational attainment, personality, and cognitive ability. Previous political science research has shown each of these traits to be strongly related to political behaviours and attitudes. For example, educational attainment is one of the strongest predictors of who participates in elections. Our project builds on this literature by testing whether genes that are associated with educational attainment, personality, and cognitive ability are indirectly related political behaviors and orientations. More specifically, we will test whether these genes influence educational attainment, personality, and cognitive ability and these traits then shape political behaviors and orientations. The scores will be derived from the Understanding Society dataset by linking individuals’ responses to questions regarding political behaviours, identity, and affiliation to their genetic data.
|Voting behaviour; educational attainment; genetics; polygenic scores||Prof Melinda MILLS||9/12/2017|
|University of Oxford||9/19/2017|
|Felix Tropf, Chris DAWES (NYU)||10/24/2017|
|National Institutes of Health|
|Characterising the genetic basis of mental health related symptoms in the general population||While most people do not have psychiatric disorders, it is possible to experience mental health symptoms that are related to some of these disorders, for example feeling low or anxious for a period of time. In the UKHLS study, participants answered a questionnaire about this kind of symptoms. We would like to study the biological background of these symptoms. Specifically, we are trying to find out how much the biological factors causing mental health symptoms overlap with the ones causing psychiatric disorders. The way we approach these biological factors is to compare the genetics between mental health symptoms and psychiatric disorders. If they overlap, it may be possible to learn more about psychiatric disorders in cohort studies that have collected genetic and mental health data, even if they have no direct information on psychiatric diagnoses. Therefore, our study can provide crucial guidance for future research to improve diagnosis and treatment of psychiatric disorders.||biomarkers, gene-environment interactions||Dr Karoline KUCHENBAECKER||7/3/2017|
|University College London||7/12/2017|
|Ka Wai (Kathy) CHAN||10/24/2017|
|Investigating the genetic relationships between anxiety, depression, stressful life outcomes, and cardiovascular risk factors and disease.||This study aims to use cardiovascular risk measurements and diagnosis, together with questionnaire data on mental and physical health in Understanding Society in two ways: 1- To discover and validate previous findings from large psychiatric genetics studies. These studies identified inherited genetic changes which may increase risk of depression. 2- To investigate shared genetic factors affecting mental and cardiovascular health (heart disease), as these conditions often occur together.
The ultimate aim is to uncover biological pathways underlying the relationship between cardiovascular disease and depression. The proposed work will be achieved through analysis of genetic and health data, using existing methods. This research will assist in risk prediction, informing treatment, and forming a better understanding of the shared genetics between traits.
|depression, polygenic risk scores, cardiovascular disease risk, pleiotropy||Prof Cathryn LEWIS||7/3/2017|
|King’s College London||7/6/2017|
|Delilah ZABENEH, Karen HODGSON, Saskia HAGENAARS, Gerome BREEN, Paul O’REILLY||8/9/2017|
|NIHR Maudsley Biomedical Centre|
|Impact of economic conditions in year of birth on DNA methylation age acceleration||Studies have found that being born in a good or bad year for the national economy can have long term effects on one’s health. For example, those born in an economic downturn experience greater risk of dying from cardiovascular disease. The social and biological mechanisms of this effect are poorly understood. One possisble explanation is that a poor economy at birth may contribute to factors (e.g. malnutrition, stress, etc.) impacting regulatory features of cells, called DNA methylation, increasing the chances of aging more rapidly. Understanding Society is ideally suited to address this knowledge gap: having DNA methylation from many adults born before 1960, when this effect is likely most powerful. We propose to investigate whether poor economic conditions at birth are related to altered DNA methylation in adults, and in particular to a measure of one’s ‘DNA methylation age’ to determine if this contributes to accelerated biological aging.||DNA methylation, age acceleration, business cycle||Prof Caroline RELTON||5/25/2017|
|SSCM, University of Bristol||6/7/2017|
|Dr Paul YOUSEFI, Prof Gerard vAN dEN BERG, Dr Matthew SUDERMAN||7/25/2017|
|ESRC and BBSRC|
|Investigating epigenetic changes in shiftwork – a possible mechanism for its impact on health and the body clock||Shift work is a widespread feature of our society, with a third of men and a fifth of women in the UK engaged in it. However, shift work has been linked with a number of poor health outcomes, including obesity, diabetes, heart disease, depression and some types of cancer. The mechanisms by which shift work might lead to these diseases are poorly understood. A major area of interest is the effect shift work, and night shift work in particular, has on altering a person’s body clock. There is information available on shift work patterns at multiple time points in Understanding Society. The availability of blood samples for some study participants provides an opportunity to look at biological changes associated with shift work, including the impact shift work might have on epigenetic modifications, which influence how our genes are turned on or off . This work will help us better understand how the occupational exposure of working shifts might become embodied in human biology, with the potential for long term health consequences.||Shift work, sleep, DNA methylation, circadian||Prof Caroline RELTON||5/25/2017|
|SSCM, University of Bristol||6/7/2017|
|Dr Rebecca RICHMOND, Prof George DAVEY SMITH, Prof Meena KUMARI||7/3/2017|
|MRC IEU, Bristol|
|A meta-analysis of genome-wide association studies of general cognitive
function in the CHARGE and COGENT consortia
|People’s differences in cognitive functions are associated with important life outcomes. Previous genome-wide association (GWA) studies have shown that these differences are partly heritable and are controlled by many genes each contributing a small effect. GWA studies of other complex traits, for example height, have shown that very large sample sizes are required to detect these small effect sizes. This study substantially increases the number of individuals (N > 100,000) to more than double that of the current largest GWA study of general cognitive function (N = 53949). This study aims to identify new genetic associations with general cognitive function and further investigate previous findings. We will also explore if genes that are associated with general cognitive function are also associated with health outcomes and physical measures of ageing, for example, walking speed.||Genetics, GWAS, general cognitive function, health||Prof Ian DEARY||2/27/2017|
|University of Edinburgh||3/7/2017|
|Dr Gail DAVIES||4/21/2017|
|Centre for Cognitive Ageing and Cognitive Epidemiology|
|Assortative mating and genetics||This project aims to use socioeconomic and genetic data from the Understanding Society. We will examine variation in spouses’ characteristics to investigate the similarity of partners to one another – this is also called assortative mating. We will examine different characteristics such as socioeconomic status (e.g., educational attainment) and health (e.g., weight and height). Assortative mating may have direct implications for the transmission of socioeconomic status and inequality across generations. Similarities between partners may be due to preferences for specific characteristics or could also capture underlying (unobservable) traits and constraints that people face when forming a relationship. In this project, we will first investigate whether we can use genetic variation (differences in family traits) among partners to better measure the similarity in their attributes. This information will help us to learn whether and how individuals value one or more characteristics in a partner. It will also help us to learn which type of characteristics matter (socioeconomic and/or health ones?). The project will then calculate measures of genetic predisposition (polygenic scores) to assess the degree of assortative mating in the partners’ characteristics.||Matching Models, Marriage, Schooling, Exclusion Restriction, Instrumental Variables, Polygenic Scores||QUINTANA-DOMEQUE, Dr Climent||2/27/2017|
|University of Oxford and St Edmund Hall||3/9/2017|
|BARBAN, Dr Nicola <male>; DE CAO, Dr Elisabetta; OREFFICE, Dr Sonia||6/2/2017|
|Inflammatory cytokines interleukin-6, interleukin-1-beta, and C-reactive protein as causal risk factors for depressive symptoms: A Mendelian randomisation study||The purpose of this study is to determine whether poor immune system functioning can cause symptoms of depression. We aim to do this by examining whether genetic variants that are associated with inflammatory factors predict a higher likelihood of having depressive symptoms, a method which reduces several types of bias. First, genetic variants that are known to be associated with specific inflammatory factors (interleukin-6, interleukin-1-beta, and C-reactive protein) would be identified and used to construct a genetic score which represents these inflammatory factors. Data from several other cohort studies including the UK Household Longitudinal Study would then be used to examine causal links between genetic scores for inflammation and depressive symptoms. Confirmation of these causal links may reveal pathways that can be targeted for the prevention and treatment of depression; the absence of causal links may shift the focus of future research onto other targets for therapy.||Inflammation; Depressive symptoms; Causality; Mendelian randomisation||CARVALHO, Dr Livia||7/14/2016|
|University College London||7/18/2016|
|Dr Joshua BELL, Dr Golam KHANDAKER||7/18/2016|
|MRC (ImmunoPsychiatry Consortium)|
|Genomics of social support, personality and cognition and their relation to mental health and cognitive ageing||Genes play an important role in shaping the social behaviour and cognition/cognitive ageing as they modulate the brain activity through molecular pathways; therefore, it can be said that genes regulate the expression of behaviour. Social support and cognition are correlated with an individual’s mental health as these social interactions require effective communication and participation. We would like to use information from the 1985 birth cohort to: (1) assess the impact and associations between social behaviour (social support) and cognition in individuals with and without symptoms of depression and anxiety, (2) perform a genetic analysis of social support and cognition within the same population by studying changes in the DNA of individuals, and (3) harmonise these data with data from different cohorts||Social support, social dysfunction, personality, cognition, GWAS, mental health, cognitive ageing||NICODEMUS, Dr Kristin||6/7/2016|
|University of Edinburgh||6/17/2016|
|Elvina GOUNTOUNA (contact), Thalia Perez SUAREZ, Kathy EVANS, Rosie WALKER, Lara Neira GONZALEZ, Daniel McCARTNEY||6/17/2016|
|No outside funding. The project will be undertaken in the laboratories of Dr. Nicodemus and in collaboration with Dr. Kathy Evans, Reader, University of Edinburgh.|
|FPLD1 and Severe Insulin Resistance GWAS||Familial Partial Lipodystrophy Type 1 (FPLD1) is a rare disorder of fat distribution with an extreme phenotype, associated with severe resistance to the glucose lowering effects of insulin. This study aims to investigate whether individuals with this disorder carry an excess of markers for known measures of “metabolic syndrome” risk (insulin and lipid measurements, blood pressure, BMI, waist-hip ratio) when compared to unaffected control participants from UKHLS. It also aims to discover whether these individuals carry any new markers that might contribute to FPLD1 disease risk, again, in comparison with unaffected control UKHLS participants.||Familial Partial Lipodystrophy, Severe Insulin Resistance, GWAS, genetic risk score||BARROSO, Dr Inês||12/1/2015|
|Wellcome Trust Sanger Institute||12/17/2015|
|Felicity PAYNE, Eleanor WHEELER & Allan DALY||12/17/2015|
|Human Genetics Working Group, Wellcome Trust Sanger Institute Core funding|
Last updated with approved projects 5/6/2018