Cancer Institute Seminar Series - Dr Giulio Caravagna

Hosted by Prof Charlie Swanton
 
A light lunch will be served after the seminar. 
 
With the current availability and resolution of cancer sequencing data, we are obtaining new measurements of tumour evolutionary dynamics across patients and cancer types. The development of the Cancer Evolution paradigm is today fuelled by the integration of molecular data from large cohorts where high-depth genome sequencing is available together with other molecular measurements (e.g., single-cell transcriptomic or epigenetics), and spatial imaging data. From the opportunity of measuring precise clonal evolution dynamics, it naturally stems a challenge with strong potential to impact both basic and translational research: can we translate "the way a tumour evolves" into a set of prognostic and predictive biomarkers? This rephrasing of a well-known problem within an evolutionary framework, poses the challenge to extract the most relevant statistical signals from cancer sequencing data. This analysis is however hindered by huge levels of tumour heterogeneity across and within patients, which we need to reconcile. In this talk, I will discuss two popular Cancer Evolution problems that can be approached exploiting ideas from machine learning and mathematical modelling, showing how these new solutions can elucidate evolutionary dynamics that can be translated into a set of powerful biomarkers.
 
Further information: Professor Andrea Sottoriva’s Evolutionary Genomics and Modelling Team
 
This seminar has been sponsored in part by the Biomedical Research Centre and Cancer Research UK