A program of research in cardiovascular genetic epidemiology
Lead Research Organisation:
University of Leicester
Department Name: Health Sciences
Abstract
High blood pressure affects around 30% of all adults in England. This program, involving researchers from the University of Leicester, and from St. George’s Medical School, aims to identify the genetic causes of high blood pressure. With my collaborators, I will try to identify regions of the human genome and the variants within them that affect blood pressure. We will screen more than half a million points in the genome where key variations occur in 1500 people from the 1958 Birth Cohort. We will build on these findings by studying some genomic regions in detail in the GRAPHIC study, involving 500 Leicestershire families who have had blood pressure measured over a complete 24-hour period. At the same time, I will work with other researchers at the University of Leicester to explore the impact of some common, but poorly-understood, types of structural variation in the human genome. We will also develop powerful new ways of combining all these types of information to identify genes that affect blood pressure. Our aim is to find variations that cause high blood pressure and these findings will improve our understanding of how high blood pressure occurs, and how it can be prevented and treated.
Technical Summary
AIMS
(1) to identify genetic variants associated with blood pressure (BP) and related cardiovascular phenotypes, using genome-wide and candidate region approaches;
(2) to develop new statistical approaches to maximise the power of genetic association studies, by incorporating different classes of phenotypic information and by utilising information on common types of genomic variation, including copy number variation.
MAIN OBJECTIVES
(1.1) to investigate the quantitative effect of candidate gene variants on mean 24h ambulatory BP;
(1.2) to investigate genetic and environmental determinants of circadian variation in 24hour BP;
(1.3) to assess the effect of genetic variants on key ECG phenotypes, including left ventricular mass and QT interval;
(1.4) to study genome-wide association with BP to inform the candidate gene approaches described above;
(2.1) to develop statistical approaches to model circadian variation in BP that utilize all available measures from ambulatory monitoring;
(2.2) to develop analytical methods that combine different classes of phenotypic information (e.g. binary and quantitative) to maximize the power of genetic studies of complex disease;
(2.3) to develop statistical approaches to appropriately model copy number variation, and to assess the implications of this kind of genomic variation on the statistical power of large-scale genetic association studies.
DESIGN AND METHODOLOGY
Objectives 1.1.to 1.3 utilise a funded collection of 2000 intensively-phenotyped individuals from 500 nuclear families representative of the general population. The family-based design ensures robustness to population stratification and permits segregation analysis, providing rational guidance to the future search for BP candidate genes. Our analytical approaches are based on generalized linear mixed models (GLMMs) fitted using Gibbs sampling in WinBUGS; these are suitable for family data, even when a proportion of subjects receive antihypertensives. Objective 1.4 will be achieved using data on 675,000 genomic markers on 1500 subjects in the 1958 Birth Cohort, using adjustments for antihypertensive treatment. I will also extend methods previously developed by our group, including GLMMs and graphical models, to utilize all available phenotypic information (2.1-2.2) and appropriately model common types of genomic variation, particularly copy number variants (2.3).
SCIENTIFIC AND MEDICAL OPPORTUNITIES
With careful study design, adequately sized studies and appropriate analytical approaches, associations between genetic variants and a range of common complex phenotypes, such as BP, will be found and replicated. Genetic associations with BP will inform understanding of the aetiological pathways underlying high BP, thereby informing preventive and treatment strategies that will reduce the risk of stroke, myocardial infarction and other complications of high BP.
(1) to identify genetic variants associated with blood pressure (BP) and related cardiovascular phenotypes, using genome-wide and candidate region approaches;
(2) to develop new statistical approaches to maximise the power of genetic association studies, by incorporating different classes of phenotypic information and by utilising information on common types of genomic variation, including copy number variation.
MAIN OBJECTIVES
(1.1) to investigate the quantitative effect of candidate gene variants on mean 24h ambulatory BP;
(1.2) to investigate genetic and environmental determinants of circadian variation in 24hour BP;
(1.3) to assess the effect of genetic variants on key ECG phenotypes, including left ventricular mass and QT interval;
(1.4) to study genome-wide association with BP to inform the candidate gene approaches described above;
(2.1) to develop statistical approaches to model circadian variation in BP that utilize all available measures from ambulatory monitoring;
(2.2) to develop analytical methods that combine different classes of phenotypic information (e.g. binary and quantitative) to maximize the power of genetic studies of complex disease;
(2.3) to develop statistical approaches to appropriately model copy number variation, and to assess the implications of this kind of genomic variation on the statistical power of large-scale genetic association studies.
DESIGN AND METHODOLOGY
Objectives 1.1.to 1.3 utilise a funded collection of 2000 intensively-phenotyped individuals from 500 nuclear families representative of the general population. The family-based design ensures robustness to population stratification and permits segregation analysis, providing rational guidance to the future search for BP candidate genes. Our analytical approaches are based on generalized linear mixed models (GLMMs) fitted using Gibbs sampling in WinBUGS; these are suitable for family data, even when a proportion of subjects receive antihypertensives. Objective 1.4 will be achieved using data on 675,000 genomic markers on 1500 subjects in the 1958 Birth Cohort, using adjustments for antihypertensive treatment. I will also extend methods previously developed by our group, including GLMMs and graphical models, to utilize all available phenotypic information (2.1-2.2) and appropriately model common types of genomic variation, particularly copy number variants (2.3).
SCIENTIFIC AND MEDICAL OPPORTUNITIES
With careful study design, adequately sized studies and appropriate analytical approaches, associations between genetic variants and a range of common complex phenotypes, such as BP, will be found and replicated. Genetic associations with BP will inform understanding of the aetiological pathways underlying high BP, thereby informing preventive and treatment strategies that will reduce the risk of stroke, myocardial infarction and other complications of high BP.