Deep Learning to mine genomics datasets

High throughput experiments that generate complex / information rich datasets are increasingly found in biology. However, to fully mine and visualise these datasets it is necessary to make use of computational methods using various bioinformatic approaches. One area that holds great promise for dealing with these high throughput biological datasets is that of deep learning. Deep learning has produced a great leaps in performance for a number of computational tasks, but has only recently started to be trialled in the biological community. Concurrent with improvements in Deep learning we are also seeing rapid improvements in technologies such as single cell sequencing, powered by large projects such as the Human Cell Atlas. This technology provides expression data for hundreds of thousands of cells which can be used to help characterise various biological systems. This can be done for whole organisms, across developmental time courses or between sub-regions of tumours. These new single cell datasets when combined with novel deep learning methods is likely lead to great advances in our understanding of cell biology and evolutionary / developmental processes. In this PhD we will make use of publicly available single cell datasets and develop novel deep learning strategies to fully utilise them for various biological questions. The research will have a focus on knowledge discovery in developmental processes and diseases such as cancer. Another focus will be to build systems capable of harnessing the large datasets even when only unlabelled data is available. Apart from biological knowledge discovery the project will also generate novel tools that could then be applied by the research community more generally to a wider range of datasets. The project will be suitable for a student with either a computational or biological background and a willingness to explore this exciting new frontier in biology.

Place of Study:
Faculty of Health and Life Sciences
Department of Biological and Medical Sciences, Oxford Brookes University

Eligibility: Home UK/EU applicants who must be permanently resident in UK/EU (or International by special exception)
Duration: Three years
Start date: Sept 2018
Value p.a.: Bursary equivalent to RCUK national minimum stipend plus fees (2018/19 bursary rate is £14,777)

Please note only EU/UK nationals/permanent residents are eligible to apply for this studentship. Please do not apply if you are not a UK/EU national/permanent resident. If you are not sure if you are eligible please contact Research Administrator, [Email Address Removed].

There is an additional requirement to undertake up to 6 hours undergraduate teaching/week during semesters and to participate in a teaching skills course without further remuneration.

For further information contact Dr Jonathan Lees [Email Address Removed]