Ground-breaking research by Perse student published in leading medical journals
2 Oct 2019
Thatcher Ference (Upper Sixth) has co-authored not one, but two articles published in distinguished medical journals after carrying out cutting-edge research on heart disease.
He has spent the last two summers as an intern at the University of Cambridge’s Centre for Naturally Randomised Trials, writing computer programs to more efficiently extract and analyse data to help try to find better ways to prevent heart attacks.
The first project saw Thatcher write a machine learning algorithm to create a genetic score that closely mimics a new medicine that lowers cholesterol and reduces the risk of heart attacks.
The information he found was used to help get approval for the new medicine with the findings being published in the New England Journal of Medicine (NEJM) in March.
This summer, Thatcher worked on a study that used genetics and computer science to estimate the combined effect of blood pressure and cholesterol on the risk of heart attack. The investigation was presented as a late-breaking clinical trial at the European Society of Cardiology (ESC) Scientific Congress in Paris earlier this month and published simultaneously in the Journal of the American Medical Association.
The study will be used by the ESC to create a new calculator for the lifetime risk of developing heart disease and form the basis of their 2021 Cardiovascular Medicine Prevention Guidelines.
Thatcher said he was “humbled” to have been involved in such ground-breaking research and to be named as a co-author on the published findings.
He said: “It was great to be able to contribute to such ideas by doing the background data extraction work. I realise some people don’t get one thing published during their career, so to have my name on two articles is amazing and I’m so grateful to have been part of this.”
Thatcher said his father, who works in medical research, had given him the impetus to use his passion for computer science in such a way.
He said: “My father used to extract data from the UK Biobank and other similar sources and it would take him weeks on end. He gave me the idea to use a programme to extract the data and it took about two minutes’ run time. It meant he could deal with the bigger picture things, while I could carry out the data extraction and analyses that he wanted me to do.”
However, Thatcher also worked on another innovative project for his Rouse Awards entry, developing a machine learning algorithm that could predict how much individuals would benefit from lowering their blood pressure and cholesterol to prevent heart attacks.
He said: “It was daunting but incredibly rewarding. I had to create my own algorithm and conduct a naturally randomised trial at each node to figure out how much risk reduction people would have, based around their genome. I applied a lot of areas of computer science and what I came up with was pretty elegant and did the job.
“This new algorithm could be the first way to use machine learning to personalise the prevention of cardiovascular disease and hopefully it could be used in different areas looking at genetic vulnerability and diseases.
“There’s still so much to learn and I still consider myself a beginner. I’m applying to read computer science at university so that I can try to develop new ways to use computing to improve the diagnosis, prevention and treatment of disease. I find it really exciting to discover new things that can potentially help people.”