Can our genes predict our likelihood of developing dementia?
Professor Valentina Escott-Price leads the DPUK/UK DRI Genetic Consortium project within the Neuroimmunology theme of DPUK's Experimental Medicine Incubator.
DPUK: What is the aim of your project?
VE-P: Our project aims to develop a framework to give people a genomic risk score to predict their likelihood of developing dementia. The genome contains all the genes for that person, so we want to be able to take someone's genome, calculate how many genetic variants they have that are linked to an increased risk of dementia and what that relative risk is, and give them a total risk score. We are expecting there to be a polygenic component, which means that certain genetic variants may only carry a small risk of dementia themselves, but when they are present in combination, they are associated with a higher risk of dementia. In particular, my team is focusing on genes that are related to neuroimmunology, the brain's immune system, because there is lots of evidence that overactivation of this system could be a mechanism driving dementia.
How are you investigating this?
We are using as many cohorts as possible within the DPUK Data Portal that collected genetic information about their participants. We first need to harmonise and unify the data in these cohorts to find a set of genetic variants that is available across all datasets. We will then impute the data, which involves calculating a value that predicts the relative risk each genetic variant has for dementia. However, we are not anticipating this to be perfect, so we will compare results from our data to existing cohorts. For example, we will use our data to predict how high a risk of dementia certain variants of the APOE gene carry and compare these figures to established datasets for this gene. We need to do this because many of the control participants (people without dementia) used in these cohorts are relatively young, meaning they may still have gone on to develop dementia, so may not actually be true controls. We will repeat this comparison with other known genetics until we are satisfied that our prediction accuracy is high. We can then move on to lesser studied genes in relation to dementia, such as genes related to neuroimmunology.
What stage are you at with the project?
It took us a while to get access to the data (because genetic information is so sensitive) and build the infrastructure required to process it. We now have permission from 11 cohorts within the DPUK Data Portal to use their data, which totals roughly 580,000 participants. We have already begun the lengthy task of processing this data – with the data collected for so many different purposes on different platforms, it needs a lot of tidying up before we can analyse it. We are now at the beginning of the analysis stage to investigate the genetic variants.
Why are genetics so useful for predicting risk?
Our brains and bodies change throughout our lives, but our genetics should stay the same. This means that genetics should be able to provide the earliest possible indication of a person's risk of dementia. This would give people who were predicted to develop dementia the opportunity to make lifestyle changes early enough to reduce that risk. In the future, people at high risk may even be able to take preventative medication in time to avoid developing dementia.
Are there any other implications of this research?
Dr Sarah Bauermeister, my two postdocs (Ganna Leonenko and Joshua Stevenson-Hoare), and I will be publishing a paper in the next year describing the protocol we have followed and the data that are now available. Once this information is in the public domain, other researchers can apply to access it to study their own research questions. We are expecting to provide information for over 40 million genetic variants, so there is a lot to investigate, and we are very open to collaboration. We are hoping that biologists will build on our work by studying the genes for neuroimmunology that we have found are linked to dementia in more detail to explore biological mechanisms that could be involved. This work could then lead to therapeutic drugs that target those physiological pathways to treat or even prevent dementia.