Discovery of causal relations of risk factors using Mendelian randomisation and genetic colocalization with UK Biobank data

Abstract

AbstractMany risk factors, disease traits and comorbidities have been associated with Alzheimer Disease (AD). However, given the long subclinical development of the disease it is difficult to establish temporal order or to rule out shared aetiologies in observational research. Novel analytical molecular epidemiology approaches can be used in AD to investigate potential causality. Mendelian randomisation analysis is an approach where genetic instruments are used as a proxy for the exposure and allow us to investigate the causality of the associations in an observational setting. By selecting genetic instruments MR can be extended to proxy a drug target which allows to mimic the effect of medications targeting that particular gene. Colocalization is another method that tests whether a single genetic variant is driving the association of a shared gene for two traits or disorders. The availability of large‐scale data in UK Biobank including genome‐wide association data (GWAS) and whole genome sequencing (WGS) with a range of questionnaire data, physical measurements, proteomics and other biological measurements and clinical phenotypes has the potential to develop new knowledge on AD aetiology using these new approaches. For example, we were able to study the association between features from MRI brain imaging with AD through the MR approach and show associations between genetically predicted brain iron accumulation in different brain areas with AD risk. In another approach, we were able to study pleiotropic genetic variants that affect both cardiovascular traits and AD and through colocalization identify genes that are likely to be causal for both diseases. Using the wealth of phenotypic data in UK Biobank we were able to further explore the effect of these genes across the phenome and exposome. Further molecular data is expected to be released in UK biobank including proteomics and enhance the study of molecular fingerprints of AD and to extrapolate this knowledge to medications that target them and to elucidate underlying disease pathways.