Identifying trait-causal mutations in UK biobank data using abundance-modified Mendelian Randomisation (AMMER)
Principal Investigator:
Professor Chris Ponting
Approved Research ID:
53116
Approval date:
February 4th 2020
Lay summary
The UK biobank's genetic data and information about individuals' health, preferences and lifestyle allow researchers to find genetic changes that predispose people to, say, a particular disease. However, finding these DNA patterns does not immediately reveal which DNA change is a cause of altered disease risk. There are many reasons why this is so, from the sheer size of the human genetic code to the fact that many DNA changes are inherited together down generations, making it hard to tell their effects apart. In this project we want to pinpoint genetic changes that have a real effect on people's risk of disease. To do so we will use detailed knowledge of the chemical machinery that reads out DNA into other molecules used by our cells. More specifically, we expect a large number of these causal DNA changes to be found in parts of the genome where transcription factors bind to the DNA. These factors are the molecules responsible for increasing or decreasing the reading out of DNA eventually into proteins. Changes to the genome in these areas will likely affect the ability of these transcription factors to bind to the DNA, which in turn leads to other changes in the body, ranging from altered risk of rare diseases to slight changes in height, hair colour, and so on. We propose to look for these causal variants by seeing how their effect on humans is affected by how available these transcription factors are in the cell. That is, in people who have an abundance of a given transcription factor, the effect of a related gene variant will also be large. By finding genetic variants where this is the case, we can make a convincing argument that these variants are truly causal of - for example - a change in disease risk. The study could directly help find causes for genetic diseases, but the tools we create as part of this study will also help us better understand how our genetic code is read out by molecular machines inside our cells. Once the mathematical and computational tools are ready for use, we expect to be able to apply them to a vast number of different applications, including on the UK biobank's data repository. Initial project work and investigations are expected to take around two years for three full-time postdocs and one PI.