Broadcast: News items

PhD Spotlight: Graeme Benstead-Hume

Graeme Benstead-Hume

PhD in Bioinformatics

At its core, my work revolves around using statistical methods and machine learning to find patterns in biological data. These patterns can often lead to insights into underlying biological processes and, specifically, to the identification of potential genetic vulnerabilities in cancers. Of these vulnerabilities, a class of genetic interaction called synthetic lethality has been of particular interest to me. A synthetic lethal pair is a pair of genes which are not essential to a cell’s survival individually but can cause cell death when a mutation knocks them out simultaneously. In cancer, often a tumour suppressor has been knocked out, but a related, druggable synthetic lethal partner is known; these interactions therefore provide the opportunity for the targeted therapy of cancer cells, while sparing healthy cells.

My most recent projects have focused on building networks from biological data, such as protein-protein interaction maps. Physical interactions link most proteins and by creating a map we can visualise these complex networks, allowing powerful analyses. I have been extracting topological features from these networks which are in turn used as parameters in machine learning models to predict synthetic lethal pairs or cancer gene dependencies identified in specific cancer patients.

These predictions can be used to better guide future drug discovery screening and ultimately lead to future novel tailored or personalised cancer therapies. The Downs Lab in the Institute of Cancer Research were kind enough to invite me to validate these predictions and after a month of work in their lab, I was very proud to discover that the majority looked promising! My findings are now being shared with the wider drug discovery community through an online database called SLorth, which we created. This database provides tools for researchers and clinicians to search for clinically relevant genes to inform targeted cancer treatments.

I originally studied BSc Artificial Intelligence at the University of Sussex. I then went on to work in industry as a developer and data analyst for marketing clients. After 15 years, I finally, and happily, quit marketing to travel to Singapore. A chance encounter then led to a bioinformatics post in Thailand for three years. This instilled in me a big love of big data and biology! The birth of my son, Harrison, then bought us home to the UK and thankfully Dr Frances Pearl gave me a chance to apply for a PhD position which was funded by the MRC.

“You get as much out of a PhD as you put in”: this was said to me during my initial PhD candidacy interview and has rung true for me. I actively look for opportunities to try ‘extra-curricular’ activities and I have really benefited from presenting my work at meetings, getting involved in Widening Participation to teach kids the bioinformatics basics and running stalls at science fairs.

The University of Sussex has always represented an excellent institute for collaboration and research which appealed given my varied background. The Genome Damage and Stability Centre is well recognised for its impactful research, this and the fact I was very fond of Brighton and the University in general having studied here previously, led me to pick Sussex as my first choice when contacting bioinformatics departments for PhD opportunities.

My road to a PhD hasn’t been especially direct but I am very happy it has bought me here as I can’t imagine a more rewarding field to be a part of. In the future, as long as I can find somewhere to keep on doing the kind of work I’m doing now I’ll be happy! I have started writing grants in the hope that I’ll be able to continue my work here at Sussex, join the Downs Lab at the Institute of Cancer Research, or in an ideal world, a bit of both!

By: Jessica Gowers
Last updated: Wednesday, 27 November 2019

Found this interesting? Share it on social media: