Applied genetic epidemiology and study design. : Nic Timpson
This programme of research concerns the contribution of genetic factors to complex traits and the use of genetic data within frameworks of epidemiological analysis allowing causal inference. I am currently the PI for the Avon Longitudinal Study of Parents and Children (www.bris.ac.uk/alspac). I also currently lead a programme within the MRC Integrative Epidemiology Unit (MRC IEU) focused on the development and application of Recall by Genotype (RbG) studies. These are currently being employed and developed into a nation-wide strategy across suitable participant collections including the NIHR Bioresource. In June 2016, I was successful in securing a Wellcome Trust Investigator Award which running for six years from September 2016. This work realises my continuing research focus on understanding body mass index (BMI) as a risk factor. With this I co-lead a work package applying RbG to questions pertinent to the aetiology of cancer risk and progression in the CRUK supported Integrative Cancer Epidemiology Programme (ICEP) and am part of the cardiovascular and translational work streams in the Bristol Biomedical Research Centre (NIHR).
Importantly, the work I have undertaken in the field of applied genetic epidemiology (through the course of the MRC Centre for Causal Analyses in Translational Epidemiology) and through the first five years of the MRC IEU have lead me to develop methods which are now being applied to focus questions. These are currently around the impact of body composition on metabolomic profile and health, cancer risk and progression and the aetiology of post cardiac surgery complications. All of these have causality at their heart and seek to use genetics to better understand health and disease.
Aims and Objectives
The original MRC IEU programme leader track set out to develop and apply Recall by Genotype (RbG): a study design where the recruitment of a sub-set of participants from an existing study, analysis of their biosamples or collection of new data is undertaken informed by measured genotypic variation. The original aims of the programme set out to develop a methodological toolkit for RbG studies apply different RbG designs to a series of exemplar studies and to use contributions from within this programme and across the unit to development of RbG and future directions.
The new and applied research programme sets out to better understand how body mass index (BMI) exerts an effect on human health and disease. In its first iteration, this new and focused programme will use metabolomics in a series of complementary study designs to unpick the relationship between the biological events flagged by differential BMI and health. The programme will update the limited existing lists of BMI driven metabolites and examine evidence for causal effects of these on disease. This activity will give insight into pathology, prediction and to flag causal targets for further examination across the diverse analysis scenarios presented by the MRC Integrative Epidemiology Unit (IEU).
The programme will address five key research questions by testing the hypotheses (h*) that:
(h1) Specific circulating metabolites are affected by BMI change
(h2) BMI has a causal effect on the human metabolite response to feeding
(h3) BMI has a causal effect on the on the human faecal microbiome
(h4) It is possible to causally map the human faecal microbiome onto the metabolome to extend the study of the microbiome
(h5) Metabolites identified through population based causal analysis of BMI, intervention studies of BMI change, the effect of BMI on metabolic response and investigation of the microbiome (“BMI-metabolites”) have a causal effect on disease risk
In “STEP 1”, BMI-metabolites are identified. In “STEP 2”, genetic instruments for BMI-metabolites are identified and used in multi-study MR analyses to assess disease risk in cross-section or as incident disease case analyses UK Biobank.
Alexandra Creavin; Childhood visual impairment and abnormalities of the optic disc: clinical phenotypes and associations with early-life and genetic factors. Started October 2015, funded by the National Institute for Health Research.
Alex Kwong; Examining genetic and environmental contributions to psychiatric disorders through multilevel modelling. Started September 2016, funded by the ESRC.
Tom Dudding; Vitamin D and head and neck cancer risk and progression: An examination of causality and mechanisms. Started September 2016, funded by the Wellcome Trust.
Thomas Battram; Will be using statistical and bioinformatics techniques, on multiple omics datasets, to analyse how genetic variation influences disease outcomes via regional changes in DNA methylation. Started October 2016, funded by the Wellcome Trust.
Simon Haworth; The utility of dental traits as predictors of health status or treatment need, gene discovery for dental traits and application in causal analyses. Started May 2016, funded by Wellcome Trust Clinical Research Training Fellowship scheme.
Sarah Watkins; Analysis of changes in DNA methylation over time in a longitudinal cohort. Focus on regions and networks of change, and the influences of environment and disease. Started October 2015, funded by the Wellcome Trust.
Kayleigh Easey; The effect of in utero exposures on offspring mental health. Started October 2015, Funded by MRC IEU.
Peter Taylor; Common variation in thyroid hormone bio-availability; effects on key health outcomes. Based at the London School of Hygiene and Tropical Medicine – Academic supervision for MSc in applied epidemiology – transferred to PhD 2013, Funded by the Welsh Clinical Academic Trainee scheme.
A Recall by Genotype toolkit
One of the core deliverables for P3 was a “toolkit” which to facilitate the undertaking of recall by genotype experiments. This has involved efforts to understand and employ genetic instruments in advanced MR settings and to provide recommendations on instrument use to the research community.
1.Corbin LJ, Richmond RC, Wade KH, Burgess S, Bowden J, Smith GD, et al. BMI as a Modifiable Risk Factor for Type 2 Diabetes: Refining and Understanding Causal Estimates Using Mendelian Randomization. Diabetes. 2016;65(10):3002-7.
2.Corbin LJ, Timpson NJ. Body mass index: Has epidemiology started to break down causal contributions to health and disease? Obesity. 2016;24(8):1630-8.
3.Hemani G, Zheng J, Wade KH, Laurin C, Elsworth B, Burgess S, et al. MR-Base: a platform for systematic causal inference across the phenome using billions of genetic associations. bioRxiv. 2016.
4.Timpson NJ. Commentary: One size fits all: are there standard rules for the use of genetic instruments in Mendelian randomization? International Journal of Epidemiology. 2016.
5.Millard LAC, Davies NM, Timpson NJ, Tilling K, Flach PA, Smith GD. MR-PheWAS: hypothesis prioritization among potential causal effects of body mass index on many outcomes, using Mendelian randomization. Scientific Reports. 2015;5:16645.
6.Burgess S, Timpson NJ, Ebrahim S, Davey Smith G. Mendelian randomization 10 years on: where are we now and where are we going? International Journal of Epidemiology. 2015;44(2):379-88.
7.Burgess S, Scott R, Timpson N, Davey Smith G, Thompson S. Using published data in Mendelian randomization: a blueprint for efficient identification of causal risk factors. Eur J Epidemiol. 2015:1-10.
Nicotine receptor status and smoking behaviour
Having established a collaborative relationship with Astrazeneca, we have now completed the recall by genotype project focused on the impact of genetic variation at the CHRNA3/5 locus and detailed smoking topography.
Ware JJ, Timpson NJ, Davey Smith G, Munafò MR. A recall-by-genotype study of CHRNA5-A3-B4 genotype, cotinine and smoking topography: study protocol. BMC Medical Genomics. 2014;15(13).
ZN804A variation and sleep
A RbG study based on variation at the ZNF804A locus and the precise measurement of neural oscillation in sleep was initiated with colleagues at the University of Bristol. In total - two overnight stays with electroencephalogram recordings, psychological testing and behavioural including multi axial actigraphy has been undertaken for participants selected on the basis of genetic variation at ZNF804A.
Hellmich C, Durant C, Jones MW, Timpson NJ, Bartsch U, Corbin LJ. Genetics, sleep and memory: a recall-by-genotype study of ZNF804A variants and sleep neurophysiology. Bmc Med Genet. 2015;16
GWAS and whole genome sequence (WGS) data to extend the possibility of RbG studies
Programme 3 has been involved in the undertaking of discovery GWAS for over 75 different traits; with >20 from separate investigations across >40 opriginal papers and with leading roles for many. This material is crucial for the undertaking of MR experiments including RbG. As an exemplar, the UK10K project was initiated with the objective of measuring whole genome genetic variation (via next generation sequencing) in ~10000 participants draw from the general population and from samples of rare disease. The direct link from this work to the undertaking of Programme 3 within the MRC IEU was in the potential to discover new, rarer frequency, genetic variants which are associated with predominanrly cardiometabolic phenotypes and which are available for the design of RbG studies. 7 flagship papers have been published (6 in 2015) including the overarching paper for the project, a specific bone health related paper, a cardiometabolic risk factor mapping paper, 3 ancilliary papers and one bioinformatics resource for accessing UK10K results and data.
The UK10K Consortium. The UK10K project: rare variants in health and disease. Nature. 2015;526:82-90.
APOC3 and targeting specific gene effects
One of the key outputs of the UK10K project was the report of an association between a rare and functional mutation at the APOC3 locus (at a frequency of ~0.2%) which exerted a large effect on fasting and non-fasting triglyceride levels. Importantly, this effect has been independently validated and seen to predict cardiovascular risk. We have been able to followup this work by undertaking an analysis within detailed collections of metabolomic data to assess the impact of this genetic variant beyond the main lipid moieties and explore the likely mechanisms of this gene effect.
Timpson NJ, Walter K, Min JL, Tachmazidou I, Malerba G, Shin S-Y, et al. A rare variant in APOC3 is associated with plasma triglyceride and VLDL levels in Europeans. Nature Communications. 2014;5.
Drenos F, Davey Smith G, Ala-Korpela M, Kettunen J, Wurtz P, Soininen P, et al. Metabolic Characterization of a Rare Genetic Variation Within APOC3 and Its Lipoprotein Lipase-Independent Effects. Circ Cardiovasc Genet. 2016;9(3):231-9.
A key strategic development for RbG is the initiation and trialing of a UK based recall by genotype network. Composed of more than 10 potential colaborating partners (with representation from the NIHR Bioresource and industry), the objective has been to generate an active network for the undertaking of recall studies which provided both the sample size required for this type of study design along with centres of phenotypic expertise for specific recall experiments.