Supervisors

Lead Supervisor- Dr David Richards, University of Exeter, Department of Phsyics and Astronomy

Co-Supervisor:

Professor Krasimira Tsaneva-Atanasova, University of Exeter, Department of Mathematics and Statistics

Professor Andrew Dick, University of Bristol, Departmenmt of Medicine

The GW4 BioMed2 MRC DTP is offering up to 21 funded studentships across a range of biomedical disciplines, with a start date of October 2025.

These four-year studentships provide funding for fees and stipend at the rate set by the UK Research Councils, as well as other research training and support costs, and are available to UK and International students.

About the GW4 BioMed2 Doctoral Training Partnership

The partnership brings together the Universities of Bath, Bristol, Cardiff (lead) and Exeter to develop the next generation of biomedical researchers. Students will have access to the combined research strengths, training expertise and resources of the four research-intensive universities, with opportunities to participate in interdisciplinary and 'team science'. The DTP already has over 90 studentships over 6 cohorts in its first phase, along with 58 students over 3 cohorts in its second phase.

The 120 projects available for application, are aligned to the following themes;

• Infection, Immunity, Antimicrobial Resistance and Repair

• Neuroscience and Mental Health

• Population Health Sciences

 

Applications open on 10th September 2024 and close at 5.00pm on 4th November 2024.

Studentships will be 4 years full time.  Part time study is also available.

Project Information

Research Theme:

Neuroscience & Mental Health

Summary

During this exciting, fully-funded PhD, you will use machine learning to automatically classify the state of microglia (the brain’s specialised immune cells). This will involve combining mathematics, computer programming and artificial intelligence with real experimental data to develop both supervised and unsupervised methods to predict microglial state. You will have the opportunity to collaborate with researchers in Exeter, Bristol, Newcastle and Leeds. This work has significant potential applications throughout biology and medicine, including in drug discovery, cancer and neurodegenerative conditions such as motor neuron disease, Parkinson's disease and Alzheimer's disease.

Project description

Background: Microglia are the resident immune cells of the brain. They adopt a wide range of phenotypes to control the brain’s immune response, including phagocytosing unwanted agents and releasing signalling chemicals to other cells in the brain. The scientific community has spent the last fifty years naively categorising microglial phenotype into just two types: M1 (inflammatory) and M2 (anti-inflammatory). However, recent work (including that by our collaborators) has led to the revolutionary idea that microglial state should instead be a
“multidimensional concept”, with a spectrum of states.

Importance: Determining how many states microglia can exist in, whether these states form a continuum and being able to predict microglial state is of fundamental medical importance. This is because microglia play a vital role in neurodegenerative disease (including motor neuron disease, Parkinson's disease and Alzheimer's disease) and cancer. Improved prediction of microglial state, particularly if this can be achieved from standard bright-field imaging, could revolutionise diagnosis of these conditions and provide a valuable tool in the search
for treatments by, for example, aiding drug screening programmes. Machine learning: The vision is that microglial state could be predicted simply from cell shape. A human attempt to do this would be timeconsuming and would be affected by unconscious bias and human error. Instead, what is needed is an automatic computational method. This is precisely what machine learning can achieve. Preliminary results in our group show that microglia can be classified with high accuracy (>93%) even using single cells. The aim of this PhD is to improve this.

Key research questions:
(1) What are the best machine learning techniques for automatically classifying microglial state?
(2) How do these optimal techniques depend on image size, imaging conditions and imperfect training data?
(3) Can the approach be optimised to run in real time and on multiple cells at the same time?
The approach: This PhD will leverage the opportunity presented by our collaborations with Dr Kate Harris (University of Leeds) and Prof Ian Wood (University of Leeds). It will employ a truly multi-disciplinary approach to study possible states of microglia. The student will undertake a cross-disciplinary PhD, including machine learning, image analysis and time-lapse imaging experiments. This approach will allow the student to learn a highly-desirable combination of quantitative and experimental skills, leading to excellent future career prospects.
Project plan and objectives: This cross-disciplinary studentship will be based within the Living Systems Institute at the University of Exeter. The student will also spend time at the University of Bristol and with our collaborators at the Universities of Newcastle and Leeds. Further, the student will join the Exeter Health Analytics network (which the first supervisor leads) to obtain a broad understanding of the role of mathematical modelling throughout human health, and will work with the Institute for Data Science and Artificial Intelligence at the University of Exeter. The project itself will include:

Objective 1: Creation of novel machine learning approaches, in particular convolutional neural networks (CNNs), to automatically classify microglial state based on our existing large data set of over 20,000 microglia. This will involve exploring a number of different data sets and CNN architectures (LeNet-5, AlexNet, VGG-16, ResNet,
Inception, Xception, Inception-ResNet, DenseNet and ResNeXt-50).

Objective 2: Culturing and imaging of the human microglial HMC3 cell line to generate further data for training of CNNs and to test the accuracy of the machine learning models. Cells will be activated with either interferon (IFN alfa-2b) or lipopolysaccharide (LPS). Various stains will be used to aid identification of the cell shape, including wheat germ agglutinin (WGA), CellMask and Actin ReadyProbes.

Objective 3: Design of image analysis software to automatically segment cells from raw microscopy images. This will be based on existing code in our groups. This will then be used to generate input data for the machine learning. Relevant techniques that will be considered include contrast adjustment, thresholding, morphological operations, edge detection, filtering, distance transforms and the watershed transformation.

Objective 4: Application of the approach to microglia in the eye. This will involve working in the lab of Prof Andrew Dick at the Bristol Medical School (THS) in the University of Bristol. This group has a microglia reporter mouse model that will be used to generate new images of microglia in the eye. The CNNs will then be retrained on this new data and the differences and similarities to microglia in the brain investigated. We have designed the project so the student will have significant scope to take ownership. This particularly applies to Objective 1 (where there are several possible machine learning approaches) and Objective 3 (with several image analysis options). However, objectives 2 and 4 can also be tailored as required. Importantly, the proportion of time spent on each objective can be adjusted and so the student will be able to balance the project to best suits their needs.

Funding

This studentship is funded through GW4BioMed2 MRC Doctoral Training Partnership. It consists of UK tuition fees, as well as a Doctoral Stipend matching UK Research Council National Minimum (£19,237 p.a. for 2024/25, updated each year).

Additional research training and support funding of up to £5,000 per annum is also available.

Eligibility

Residency:

The GW4 BioMed2 MRC DTP studentships are available to UK and International applicants. Following Brexit, the UKRI now classifies EU students as international unless they have rights under the EU Settlement Scheme. The GW4 partners have agreed to cover the difference in costs between home and international tuition fees. This means that international candidates will not be expected to cover this cost and will be fully funded but need to be aware that they will be required to cover the cost of their student visa, healthcare surcharge and other costs of moving to the UK to do a PhD.  All studentships will be competitively awarded and there is a limit to the number of International students that we can accept into our programme (up to 30% cap across our partners per annum).

Academic criteria:

Applicants for a studentship must have obtained, or be about to obtain, a first or upper second-class UK honours degree, or the equivalent qualification gained outside the UK, in an appropriate area of medical sciences, computing, mathematics or the physical sciences.  Applicants with a lower second class will only be considered if they also have a Master’s degree. Please check the entry requirements of the home institution for each project of interest before completing an application. Academic qualifications are considered alongside significant relevant non-academic experience.

English requirements:

If English is not your first language you will need to meet the English language requirements of the university that will host your PhD by the start of the programme. Please refer to the details in the following web page for further information https://www.exeter.ac.uk/study/englishlanguagerequirements/

Data Protection

If you are applying for a place on a collaborative programme of doctoral training provided by Cardiff University and other universities, research organisations and/or partners please be aware that your personal data will be used and disclosed for the purposes set out below.

Your personal data will always be processed in accordance with the General Data Protection Regulations of 2018. Cardiff University (“University”) will remain a data controller for the personal data it holds, and other universities, research organisations and/or partners (“HEIs”) may also become data controllers for the relevant personal data they receive as a result of their participation in the collaborative programme of doctoral training (“Programme”).

Further Information

For an overview of the MRC GW4 BioMed programme please see the website www.gw4biomed.ac.uk

Entry requirements

Academic Requirements

Applicants for a studentship must have obtained, or be about to obtain, a first or upper second-class UK honours degree, or the equivalent qualification gained outside the UK, in an appropriate area of medical sciences, computing, mathematics or the physical sciences. Applicants with a lower second class will only be considered if they also have a Master’s degree. Please check the entry requirements of the home institution for each project of interest before completing an application. Academic qualifications are considered alongside significant relevant non-academic experience.

English Language Requirements

If English is not your first language you will need to meet the English language requirements of the university that will host your PhD by the start of the programme. Please refer to the relevant university website for further information.  This will be at least 6.5 in IELTS or an acceptable equivalent.  Please refer to the English Language requirements web page for further information.

How to apply

A list of all the projects and how to apply is available on the DTP’s website at gw4biomed.ac.uk.  You may apply for up to 2 projects and submit one application per candidate only.

Please complete an application to the GW4 BioMed2 MRC DTP for an ‘offer of funding’.  If successful, you will also need to make an application for an 'offer to study' to your chosen institution.

Please complete the online application form linked from our website by 5.00pm on Monday, 4th November 2024.  If you are shortlisted for interview, you will be notified from Friday, 20th December 2024.  Interviews will be held virtually on 23rd and 24th January 2025.

Further Information

For informal enquiries, please contact GW4BioMed@cardiff.ac.uk

For project related queries, please contact the respective supervisors listed on the project descriptions on our website.

Summary