Abstract

This talk will present important biomedical application of statistics, and research in collaboration with experts in gene editing (experiments that make changes to DNA ), cancer, and embryonic development from the Wellcome Sanger Institute, the University of Cambridge, and Gothenburg University. It will present applications to gene editing and genomics of a wide range of statistical modelling and inference techniques including mixture modelling, mixed models, linear and non-linear methods of dimensionality reduction, variable selection for high-dimensional data, topic modelling, and non-parametric hypothesis testing. I will present scSNV-seq, an approach to combine gene editing with both DNA and mRNA abundance (measure of gene activity) readout for single cells (Cooper, Coelho, Strauss et al., 2024). scSNV-seq allows us to link gene editing outcomes, which can be generated to reproduce DNA changes in e.g. cancer patients, to changes in how the cell functions. In addition, I will be presenting research on complex developing cell populations, presenting a suite of statistical tools designed to understand the role of perturbations of particular genes in development or the effect of perturbations by disease in cancer samples (Strauss et al., 2023).