Computational Scientist in Machine Learning Genomics
I am a post-doctoral researcher working with Jim Hughes (MRC Molecular Haematology Unit, MRC WIMM). My research on the interface of machine learning and functional genomics focusses on understanding gene regulation in its three dimensional context.
Although general principles of gene regulation are well described, the role of the spatial context and the precise mechanisms and consequences of sequence variations in the regulatory genome are poorly understood. This is problematic, because the vast majority of disease associated single nucleotide polymorphisms identified through genome-wide association studies are actually located in regulatory rather then coding regions. Furthermore, assessing regulatory variations has long been neglected in research and treatment of various diseases, including cancer, owing to our lack of ability to identify them and to interpret their functional consequences.
I am particularly interested in deep learning techniques and their application for dissecting the regulatory context of enhancers and insulator elements and deciphering their precise regulatory code through which they act. Based on that I am developing more sophisticated models for predicting the impact of regulatory sequence variations in the 3D genome. These help to promote the consideration of regulatory variations in translational research and patient care and streamline methods for assessing and targeting them.
DeepC: predicting 3D genome folding using megabase-scale transfer learning.
Schwessinger R. et al, (2020), Nat Methods, 17, 1118 - 1124
Sasquatch: predicting the impact of regulatory SNPs on transcription factor binding from cell- and tissue-specific DNase footprints.
Schwessinger R. et al, (2017), Genome Res, 27, 1730 - 1742