Machine Learning regulation at the 3’ UTR
Dr Crispin Miller, Computational Biology
The 3' Untranslated Region (UTR) of a transcript presents a complex set of binding sites that are used to regulate both transcript stability and downstream protein translation. These include target sites for microRNAs (miRNAs) as well as a multitude of RNA binding proteins. These act together in tightly coordinated patterns in order to control protein expression. Many of these systems are perturbed in tumours, resulting in aberrant protein levels. A detailed understanding of how 3' UTRs interact with this regulatory machinery is therefore critical to our understanding of how post-transcriptional events control tumour growth and maintenance.
The goal of this PhD is to use Machine Learning to investigate regulatory patterns within the 3' UTR, and to ask how they help coordinate genome-wide expression programmes. The project will involve working closely with bench-scientists within the RNA and Translation Control in Cancer Group, who are using high throughput approaches to profile the regulatory interactions that occur at the 3' UTR. It will therefore involve building multi-omics models that integrate genome-wide deep sequencing and protein mass spectrometry datasets with single cell sequencing.
The studentship will be based in the Computational Biology Group and will have access to substantial computing hardware, including GPU-accelerated machine learning systems. While detailed knowledge of cell biology would be advantageous, it is not a pre-requisite, and this project is particularly well-suited to enthusiastic scientists from numerical disciplines who are interested in applying their computational skills to fundamental questions about cell biology, and to ask how these advance our understanding of tumour biology.
For informal enquiries, please email Dr Crispin Miller (C.Miller@beatson.gla.ac.uk)
To apply, please complete a PhD Studentship Form (in right-hand menu opposite). Application deadline: 3rd January 2020