Speaker: Daphne Ezer (Alan Turing Institute & Warwick)

Title: Open challenges in privacy-preserving genomics research

Abstract: Genomics research can yield valuable medical information -- it can help a doctor predict a patient's risk of developing a disease and can help identify specific genetic variants that may cause diseases, and this might lead to a possible treatments. However, genomes encode sensitive information related to personal traits, disease risk, and family relatedness, so it is important that research can be performed in a way that preserves the privacy of study participants. In this talk, we will describe privacy concerns arising from genomics research, namely how genomic information is collected, shared across organisations, and analysed. Specifically, we will dive in depth into two case studies: i) A case in which multiple different genomic data sources wish to pool their data in order to identify specific genetic variants that are linked to disease risk and ii) A case in which a private direct-to-consumer genetic testing company aims to build a predictive model of disease risk without revealing information about the data set used to train the model, while also preserving the privacy of the consumer. We will review different strategies that have been used to address these different problems, and discuss open challenges.