As many know, I am now a student at the University of Oxford reading for a DPhil (Oxford’s PhD) in Genomic Medicine and Statistics. This is a doctoral training centre (DTC) program that lasts 4 years instead of the standard 3 year DPhil that Oxford offers. It is funded by the Wellcome Trust. With the added year I have the opportunity to take a term of taught modules followed by two 11 week rotations.
Now that I’ve completed the first term and had the holiday to decompress from it I wanted to give my initial impressions of the program.
First, the cohorts for this program are incredibly small and diverse. There are ~5 students at the start of the Fall term (Michaelmas) for any given cohort. The backgrounds of everyone are incredibly diverse. Myself, I have more of a computational skillset coming into the program, but others have more of an experimental background. My cohort is unquestionably dominated by computational individuals.
The taught modules for this course are incredibly brief, but rewarding. While the standard statistics, scripting (R, Python, and bash), and research ethics courses are a part of this term they are brief and not the focus. More importantly, this term offers students the opportunity to gain insight and practical experience with a number of faculty and students within the Medical Sciences Division (the larger department that encompasses most of the campus that Wellcome Centre for Human Genetics sits on).
These modules, similar to the cohorts, are diverse. They range from cell imaging and manipulation with Dr. Catherine Green to deep learning with keras and tensorflow with Dr. Gerton Lunter. Of course, there is also a great deal of focus on current “next generation”
sequencing technologies and genome wide association study techniques. A benefit of this program is the exposure to the large consortia that are striving to gather sequence and clinical information across the UK and wider, such as the 100,000 Genomes Project, PCAWG, and WGS500.
One of the areas of particular interest to me was single cell sequencing and expression technologies, which I feel will be transformative to our understanding of cancer intra-tumor heterogeneity in a similar fashion to multi-region WES/WGS has. However, the amplification techniques used are still under development and handling allelic dropout when capturing single cell expression is still problematic.
The strengths of this program are really the ability to do an experimental and a computational rotation at the end of the taught modules (if you so choose, although there is talk of making this compulsory). I am currently preparing for my first rotation with Dr. Gerton Lunter and Dr. David Wedge with whom I will be, vaguely, working on applying deep learning techniques to regulatory elements in the human genome and then performing an analysis on the PCAWG dataset. My second rotation will be with Dr. Simon Leedham and we are still in discussion over the project.
The program, relatively still in its infancy, is changing each year. Our cohort has provided our feedback to the administrators in an effort to help continue to build and improve this program.
I hope to provide some updates regarding the rotations upon completion and some more personal blog posts about my experiences in Oxford.