Alzheimer?s Disease and Immune Events

Our aim is understanding the role of the immune system in cognition and Alzheimer?s disease (AD). To reach this goal, we want to analyse how a number of exposures (genes and events of the immune system) affect AD outcome (and cognitive function as a proxy), and then identify the mechanisms of these effects (e.g. brain morphology, circulatory system). Some of our exposures (e.g. CLU, CD33, DKK1, NSAIDs) have a proven (but not understood) link with AD, other exposures are related events (e.g. cancer, oral infection). Our target mechanisms are all potential stages where this link materialises. We think understanding the role of the immune system in AD is the essential step towards understanding AD. The best approach for this is to systematically investigate how immune genes and events affect AD and cognition, and UK-Biobank contains the essential variables required for such a systematic investigation (i.e. simultaneously recorded genes and medical history). Our investigation will render results closely matching UK-Biobank purposes: (1) understanding the role of the immune system in AD; (2) finding novel early immune biomarkers of AD; (3) developing statistical tools to better analyse large and multimodal medical datasets. As an informatics team, we have a range of skill sets and we plan to divide the research accordingly. Participant clinical measurements will be mined by Dr Chi-Hun Kim with a particular focus on linking immunologic events with cognition and multimodal outcomes. Genomic data will then be compared to relevant characteristics at a genome-wide level and focusing on immune gene sets of interest by Dr Laura Winchester. Overseeing and machine learning analysis will be implemented by Dr Alejo Nevado-Holgado. We intend to implement a full cohort analysis. From the full cohort, the effects of immune genes and events will be tested on cognitive tests and the diagnosis and onset of AD. We will approach this question from four fronts: GWAS (all SNPs needed), PheWAS (most phenotypes needed), pathway analysis and factor analysis (to investigate the causal factors in the multilevel sense). For this reason, we think it is vital to have access to the full UK-Biobank cohort with set variables.