University of Hong Kong Featured PhD Programmes
University of Southampton Featured PhD Programmes
University of Edinburgh Featured PhD Programmes

MRC DTP: Accessible machine learning in behavioural neuroscience to address the 3Rs challenge

School of Life Sciences

This project is no longer listed on and may not be available.

Click here to search for PhD studentship opportunities
Dr R Langston , Dr S Martin No more applications being accepted Competition Funded PhD Project (Students Worldwide)

About the Project

Behavioural neuroscience using rodent models is an established and essential area of biomedical research, but one which can be high cost in terms of time, finances and animal welfare. At the forefront of modern animal research is the need to address the principles of the 3Rs- Replacement, Reduction and Refinement. New state-of-the-art techniques and apparatus have improved experimental design and data collection and hold the potential to allow huge amounts of information to be extracted from minimal numbers of animal subjects. Refinement of the experimental process and reduction of the number of animals used are standard requirements for most behavioural neuroscience labs. Accessible machine learning tools however are not, but importantly this type of approach holds the ability to independently model new hypotheses based on pre-existing data. Improving availability of these tools could replace experiments and animals that would otherwise not produce useful results.

This studentship will seek to address principles of the 3Rs by:

Developing new machine learning models to investigate and characterise data.
Packaging machine learning algorithms and data analysis libraries in a form that is more readily accessible to a bench scientist, allowing disparate data to be integrated.
Identifying potential new biological insights from reanalysing legacy data.
Developing new animal handling approaches that allow the collection of more naturalistic data from minimally stressed subjects.
The Langston lab and collaborators have a large archive of electrophysiological and behavioural data covering a broad range of neurobiological investigations. In addition, the acquisition of new data can be automated, especially regarding animal handling. The effect of animal handling on performance in behavioural tests where animals are freely moving has not been rigorously evaluated regarding replicability in the context of handling stress. Over the last decade microelectronics developments have allowed low cost development of mechanical automation that can be leveraged to build test environments that reduce exposure to stress in animal subjects. We have strong collaborations with equipment manufacturers in this area which could be an ideal complementary strand to the PhD project for an appropriate candidate.


For more background to the project please see the following references:

FindAPhD. Copyright 2005-2021
All rights reserved.