We develop technologies for doctors to make better decisions faster

We tackle some of the most deadly cancers with genomes that look like a chaotic mess of missing and redundant parts – like ovarian, oesophageal and pancreatic cancer.

We work on (1) understanding the mechanisms behind this genomic chaos; (2) how to target cancer weaknesses with therapies; (3) how to overcome resistance to treatment; and (4) how to spot cancers as early as possible.

The methods we use combine machine learning and AI with experimental techniques ranging from single cell sequencing to tissue imaging. Some highlights of our research:

The 17 deadly CINs

We have developed a compendium of 17 copy number signatures characterising different types of chromosomal instability (CIN) and continuously work on extending and refining it. Current efforts are to further validate the signatures in large data cohorts and to extend them to single cell data, which will allow us to measure currently active mutational processes and evaluate how well they predict treatment response. This work is funded by an ERC (now UKRI) grant and the foundation of Tailor Bio, a genomics start up.

Read more about copy number signatures: Drews et al (2022), Macintyre et al (2018); and their applications: Smith et al (2023), Vias, Morrill et al (2023), Cheng et al (2022), Jiménez-Sánchez et al (2020).

Spotting cancer in the blink of an AI

We are creating AI approaches to analyse data from the Cytosponge, a minimally invasive device to detect a precursor of oesophageal cancer. Our work automates labour-intensive tasks and refines pathology biomarkers to identify the patients at the highest risk of developing cancer.

Read more about it: Gehrung et al (2021), Berman et al (2022)

Come and shine!

Our supportive and interdisciplinary lab is a great incubator for your career, wherever it takes you.

Selected publications