Approved Research

Developing a transethnic polygenic risk score for Alzheimer's disease

Lay summary

Our understanding of the genetic basis of common diseases, like asthma, type 2 diabetes or Alzheimer's disease, has improved hugely over the past decade since the human genome was first sequenced. We now know that most complex diseases are not caused by a single malfunctioning gene such as is the case with an inherited condition like cystic fibrosis but are due to numerous genetic variants spread across the genome each conferring a small risk to the overall chance of developing a condition. Because the risks conferred by these genetic variants is low, individually, they are not very predictive of the disease risk. What is required, therefore, is a new approach that considers the aggregate of all the small risks that a person carries in their genome.  Such approaches are now being developed, methods known as 'polygenic risk scores' or PRS are used to generate a score from the sum of the risk variants carried by any given person. These score can then be used to identify individuals who are at high risk of developing a disease. In our case, we are developing a score that will identify individuals that are at high risk of Alzheimer's disease before the onset of any symptoms. While our score appears to work well across a broad range of populations tested, including East Asian, South Asians, Africans and Europeans, in order for us to finalize the the score we are required to test it in a large sample such as the UKB. We are therefore requesting access to the UKB data and the proxy dementia phenotype that has been used for similar studies to ours in order so we can determine how well our new AD score predicts Alzheimer's disease onset.