Funded TREE projects - Spring 2020

Seed corn projects run for six months. Spring 2020 (Round 7) projects are listed in the drop-down panes below.

Lead Academic Co-Investigators Centre Fellow(s) Project title 
Jonathan Ball (CLES/Biosciences)

Rod Wilson (CLES/Biosciences)

Krasimira Tsaneva-Atanasova (CEMPS/Mathematics)

Cosima Porteus (CLES/Biosciences)

Nicolas Pugeault (CEMPS/Computer Science)

Arran Hodgkinson

Estimation of volume and morphological abnormalities in larval zebrafish using machine learning.


Animals are used to determine both the effectiveness of a drugs and to investigate any side effects which could occur and the impact of changing environment. The fresh water larval zebrafish has the potential to be used for understanding both drug and environmental impact. The zebrafish has increased its role in drug discovery towards human trials, investigations for analysis and large numbers are being used for this as an alternative species to large numbers of mammals (i.e. rats and rabbits).  

Measurements of length and weight to determine growth are good indicators of overall health of animals including fish but it’s hard to measure in very small, fragile animals. The larval zebrafish (fragile and <4mm long) is used extensively in research for screening of new drugs as well as for determining possible toxic nature of different chemicals/conditions (increased temperature due to global warming etc) found in our environment. Reduction/Replacement are overarching aims to make the best use of any animal work, this means using as much data that is available from each individual animal to reduce the number of animals used in any experimentation.

We have identified a need for a computer program that would help researchers working in these fields to: 1) determine the volume (how big) zebrafish larvae are by using pictures of these animals that can be taken easily and non-invasively; and 2) determined if the fish are showing any structural defects (that requires many years of experience and training and is time consuming). This will be achieved using combined expertise in zebrafish biology and programming skills and machine learning to develop software that will be able to do these decisions rapidly.  The work will create automated way of scoring the features of the zebrafish for example the body shape and length of the animal, and therefore remove an individual’s interpretation of these features and increase repeatability. The use of computer analysis to allow the generation of more data from existing experiments, faster and more accurately could reduce the number of experimental animals required.

Once this model has been established for one species (zebrafish here) then it can be adapted to other species of fish. The use of image analysis in aquaculture would help identify stressors (as an early warning system) before symptoms of ill health are observed (e.g. by measuring growth rates). This means actions to improve fish health could be implemented before severe health impacts are seen.

This tool will be useful biologists and users of the zebrafish in research to quickly and noninvasively assess potential drug structural side effects, on growth, fitness and health of fish and to assess toxicants in a repeatable and reliable manner.

Lead Academic Co-InvestigatorCentre FellowProject title 
Gordon Brown (CLES/Biosciences)
Ryan Ames (CLES/Biosciences)

Brandon Invergo

Understanding the cellular functions of MelLec, a novel receptor in anti-fungal immunity


Fungal infections kill more than 1.5 million people every year. Most of these deaths occur in individuals whose immune systems are altered. Understanding how our immune system deals with fungal infection is essential if we are to develop new therapies to overcome these diseases.

Our laboratory has discovered how the immune system “sees” (recognises) fungal organisms, and we have shown that these recognition systems are essential to protect against fungal infections. We have recently identified a completely new mechanism of recognition that relies on a specific molecule on the surface of cells, which we have shown to be essential to protect against some fungal infections in people. Importantly, we have no idea how this molecule mediates its defensive activities. Understanding these activities will tell us how it provides protection against these infections.

This project aims to gain these insights by using previously generated biologically-derived datasets that can be analysed using advanced computer analyses (called bioinformatics) that requires specialized expertise. In addition to providing important insights into an exciting new area of research that is of direct relevance for human health, this project will bring together expertise in infectious disease biology, bioinformatics and big data.

Lead Academic Co-Investigators Centre FellowProject title 
Priyanka Dey (CEMPS)

Eros Mariani (CEMPS)

Francesca Palombo (CEMPS)

Douglas Ferguson (NHS)

Nick Stone (CEMPS/NHS)

Ben Sherlock Modelling optical response of colloidal plasmonic nano-assemblies for use in cancer diagnosis


In the UK, 1 in 2 people will be detected with cancer in a lifetime and though the survival rates have doubled since the 70s, it is still only 30-50% on average across various types of cancer. This needs to be improved to ensure healthy life to the ageing population. Cancer survival rates can be improved if it is detected at a stage where the tumour growth can be treated efficiently. Thus precise and effective detection methods are the key.

Present cancer detection often involves extracting tissue from a suspected cancer region using a long-needle and is called a “biopsy”. This is not only an operative type of procedure but also needs days to weeks to receive the results after the tests has been completed and analysed. Other full-body scan techniques like MRI need a contrast agent to be injected into the blood which then highlights or provides contrast to make the tumour visible in the scan. Unfortunately, there has been a big concern about the toxic health effects of the MRI contrast agents, as well as due to magnetic radiation effects of the instrument. Additionally, they are expensive tests and is not usable by the doctors in the operation theatre, if required. Hence, our interest is in developing cancer detection technique using light, which is not harmful like radiation, provides quick test and analysis within minutes, is cheaper and has the ability to be combined into set-ups for in-operation use.

We are interested in developing Raman spectroscopy as a cancer detection technique using light. To improve the efficiency of the detection often
specifically synthesized gold nano-structures, which are essentially small beads of gold, need to be used. These gold nano-structures are not toxic and once injected, floats in the bloodstream and reaches the cancer tumour and is deposited into it. Light can then be used as a scanning wand onto the patient and the presence of a specific Raman signal with the receiver indicate the presence of the cancer tumour in that location.

The gold nano-structures importantly help in concentrating the light and maximizing the Raman signals for better detection of the tumour. Carefully designed nano-structures, often assemblies of small nanoparticles i.e., nano-assemblies, act as Raman signal enhancers. The light concentration is primarily dictated by the nano-assembly morphologies (the way these individual nanoparticles are arranged) and there is a great need to optimize the morphology with respect to various physical parameters to be able to maximize the light concentration, the Raman signal and thus the detection efficiency. Experimentally achieving this is labour-intensive, whereas modelling of the light concentration or optical response of nano-assemblies could help us narrow-down our search for the optimum nano-assembly morphology.

In this project, we thus propose to create theoretical models of nano-assemblies combining the team’s expertise in chemistry, physics and mathematics, then validate the developed model with our experimental data. This knowledge will then be able to predict suitable nano-assembly morphologies. The optimized nano-assembly morphology will later be prepared in the laboratory and actually tested for its detection efficiency.

Lead Academic Co-Investigators Centre Fellow(s) Project title 
Nicholas Kennedy (NHS/UEMS)

Simeng Lin (NHS)

Matt Anderson (TREE)

Tariq Ahmad (NHS/UEMS)

James Goodhand (NHS/UEMS)

Brandon Invergo A multiomic approach towards understanding anti-TNF treatment failure in patients with Inflammatory Bowel Disease


Crohn’s disease affects at least 115,000 people in the UK and millions worldwide. Symptoms include abdominal pain, diarrhoea and weight loss. These are often severe.  

Anti-TNF drugs are effective medicines. They work by blocking inflammation in the gut. Unfortunately, not everyone responds to these drugs.  Many people that do respond eventually lose response.

Currently, a limited number of alternative therapies are available. Many people face surgery as the next option. Therefore, it is important to understand why some people do not respond to anti-TNF treatment.

We will analyse blood samples from the “personalised anti-TNF therapy in Crohn’s disease (PANTS) study”. We will use the latest techniques to understand the relationship between a person’s genes, environment, and the proteins they produce. We will compare the people who have responded to anti-TNF drugs and those who did not.

By better understanding success and failure of anti-TNF treatment, doctors can target the safest and most effective therapy for individual patients. Finding important pathways involved in anti-TNF treatment failure may also help develop new drugs.

Lead Academic Co-Investigators Centre FellowProject title 
Luke Pilling (UEMS)

Joel Tabak (UEMS)

Karen Knapp (UEMS)

Lucy Banfield (UEMS)

Professor David Melzer (UEMS)

Dr Janice Atkins (UEMS)

Ben Sherlock

Identifying haemochromatosis patients from DXA images of bones and joints using deep learning methods.

The most common inherited disease in northern Europeans is known as haemochromatosis, which is caused by mutations to a gene that increases the amount of iron absorbed from the diet. Sufferers of this iron-overload disease build up iron in their organs, and in later life suffer serious diseases such as liver disease and osteoarthritis. Identifying patients early would mean that preventative treatment could be put in place.

We propose to analyse x-ray images of the bones of haemochromatosis patients using computational approaches. Features that help to identify people with iron-overload can then be used in the clinical setting as a screening tool: x-rays are commonly taken for all sorts of reasons, and we could therefore use data already available in the NHS to identify and treat haemochromatosis patients early. This would substantially reduce disease in these patients.




Lead Academic Co-Investigators Centre Fellow(s) Project title 
Genevieve Williams (CLES)

Stana Zivanovic (CEMPS)

Vicki Goodwin (UEMS)

Piotr Slowinski Maintenance of Symmetry and Stability in Human Walking

Walking is part of our day-to-day activity. If we get injured, for example sprain our ankle, or we are unwell with a condition that impairs our movements, for example after a stroke, our walking is affected. This has a huge impact our everyday lives. To recover our walking, we will often receive physiotherapy.

Current physiotherapy practices encourage us to walk in a symmetrical way, that is, our left and right sides do the same each other. However, we have reason to believe that if we focus on stability of walking, rather than symmetry, the rehabilitation might be more effective for helping you to walk in a way that is functional to you.

To test this, we have broken the natural body symmetry to replicate an injury or illness that effects one side of your body. We have then measured what healthy-young people, people who are older, people on unsteady surfaces, and people who are really tired do; so that we can understand the importance of symmetry and stability in walking.After this project, when we know the answer, we will look to improve physiotherapy for walking, in order that you can walk more effectively after injury or illness.