Cookies on this website
We use cookies to ensure that we give you the best experience on our website. If you click 'Continue' we'll assume that you are happy to receive all cookies and you won't see this message again. Click 'Find out more' for information on how to change your cookie settings.


About the Research

We are interested in building computational models to understand and predict the functional consequences of mutations in the (human) genome.  The crucial ingredients that we use are data, and machine learning tools to learn patterns from these data.  We apply machine learning approaches that help to inform hypotheses about how biology functions, and help to assess mutations that lead to clinical hypotheses.  We collaborate extensively with the group of Jim Hughes who is interested in the same scientific problems; within this collaboration our focus is on the machine learning and statistical aspects.

We currently run projects that look at aspects of building "deep learning" computational models to predict the consequences of mutations on splicing; chromatin accessibility; 3D chromatin interactions; and gene expression.  Broadly our direction is towards more integrated models that capture several of these aspects simultaneously.  Our data sources are in principle any genome-based assay, with a strong focus on chromatin accessibility (ATAC-seq), gene expression (RNA-seq) and chromatin interaction data (Hi-C and capture-C). 

Increasingly we are interested in analysing single-cell data which promise a much more fine-grained few of tissue diversity, across the tissue type, developmental, and cell state axes.  Analysing this data is challenging because by its nature single-cell assays are noisy.  We are interested in developing and improving the computational tools for single-cell analyses, which involves developing bioinformatic and statistical analysis pipelines.

Informal enquiries are welcomed and can be directed to

Training Opportunities

Students are expected to have good quantitative and computational skills and/or a strong desire to build on their existing interests in this direction.  Depending on the project and interest there will be opportunity to apply state-of-art machine learning models to large data sets, including working with Tensorflow/Keras and GPUs, as well as applying and developing probabilistic models.

Students will be enrolled on the MRC WIMM DPhil Course, which takes place in the autumn of their first year. Running over several days, this course helps students to develop basic research and presentation skills, as well as introducing them to a wide-range of scientific techniques and principles, ensuring that students have the opportunity to build a broad-based understanding of differing research methodologies.

Generic skills training is offered through the Medical Sciences Division's Skills Training Programme. This programme offers a comprehensive range of courses covering many important areas of researcher development: knowledge and intellectual abilities, personal effectiveness, research governance and organisation, and engagement, influence and impact. Students are actively encouraged to take advantage of the training opportunities available to them.

As well as the specific training detailed above, students will have access to a wide-range of seminars and training opportunities through the many research institutes and centres based in Oxford.

All MRC WIMM graduate students are encouraged to participate in the successful mentoring scheme of the Radcliffe Department of Medicine, which is the host department of the MRC WIMM. This mentoring scheme provides an additional possible channel for personal and professional development outside the regular supervisory framework. The RDM also holds an Athena SWAN Silver Award in recognition of our efforts to build a happy and rewarding environment where all staff and students are supported to achieve their full potential.



R Schwessinger,M Gosden,D Downes, R Brown, J Telenius, YW Teh, Gerton Lunter, JR Hughes. DeepC: Predicting chromatin interactions using megabase scaled deep neural networks and transfer learning. bioRxiv 10.1101/724005

R Brown, G Lunter. An equivariant Bayesian convolutional network predicts recombination hotspots and accurately resolves binding motifs.  Bioinformatics 35(13), pp. 2177-2184

D Cooke, DC Wedge, G Lunter. A unified haplotype-based method for accurate and comprehensive variant calling. bioRxiv 10.1101/456103


MRC WIMM Centre for Computational Biology© Martin Phelps





Part of the Centre for Computational Biology

Find out more