Spring 2015 Projects

The causes of type 1 diabetes are poorly understood, making both timely diagnosis as well as identification of new treatments very challenging. There is currently no cure for the disease and treatment options rely on the regular administration of insulin. In this project, we will investigate how the disease develops by considering the changes happening within the pancreas early in the disease course. New research has shown that there are two forms of type 1 diabetes, one which is aggressive and rapid in onset, with the second having a slower progression. These two forms of the disease may respond differently to the same treatment plan, so it is important to understand the mechanisms distinguishing them. 

Using powerful mathematical techniques, we will construct a virtual environment in which to explore and analyse the disease progression under a variety of conditions, accounting for variability between patients. By including previously overlooked details, we will investigate how immune cells work together to mount a coordinated attack on the pancreas and explore ways in which the destruction of the insulin secreting cells might be reduced.

The yeast Saccharomyces cerevisiae, better known as bakers/brewers yeast, is found naturally on decaying fruit and is used widely in winemaking, baking and brewing. Another yeast species, Candida glabrata, is a common cause of oral and vaginal infections in humans. Despite these organism’s drastically different lifestyles they are very closely related. They share the vast majority of genes and these genes even have the same order along chromosomes. We wish to understand how a species very similar to bakers yeast is able to cause disease in humans. One possible explanation is that in these species, genes are turned on and off in different combinations under different environmental conditions. To identify these differences in gene expression we will examine the expression profiles of these species under a variety of conditions. Using a new method to represent gene expression data we will identify sets of genes associated with causing disease. The outcome of this work will be the discovery of gene expression patterns that have allowed a species to become a pathogen of humans. The tools employed in this work and its findings will be important to understand how yeast causes disease and also identify potential targets for disease treatment.

In diabetes, sulphonylurea (SU) treatment is one of the first line drugs to be used, which improves glucose control by stimulation insulin release from pancreatic beta cells. However, some patients find that their blood sugar control eventually worsens despite SU treatment. This reduced response varies between patients, with some taking a long time to “fail” on SU, whilst others benefit for only a short time. Typically, those patients that respond well to SU treatment also fail the fastest, suggesting that the response to the drug drives the “failure”, but the mechanisms for this are not known. Our data suggest that the biological target of SU drugs, the K_ATP­ channel, adapts to SU treatment, which may cause the reduced efficacy of medication. We will use experimental results to develop a simple mathematical model to predict the changes in K_ATP channel behaviour following drug exposure. If we are successful, we may be able to predict the time to failure for each patient and adjust consecutive treatment accordingly.

The shape of organelles is essential to the function of mammalian cells. Organelles are small membrane-bound compartments within cells that play vital roles in the function and maintenance of all cell types. Our biological work has for some time focused on a particular organelle, the “peroxisome”, responsible for a whole host of important reactions including breakdown of fatty acids, lipid synthesis combating oxidative stress. The number and shape of peroxisomes is not fixed, but continually change as need requires. In particular, the number of peroxisomes can increase in a fascinating process, where the organelles sprout long tubular extensions that subsequently divide into small, new peroxisomes. 

In this project, we will try to understand this process and look it detail at what controls the organelle shape during extension and division. This has not been done before and will involve combining high-quality microscopy images with mathematics and computing. 

Understanding this process is important since various human patients show abnormalities in peroxisome growth and division. This work will help to identify such patients and to propose treatments, something that at present is almost entirely lacking.

The attachment of chemical tags to DNA (known as DNA methylation) can affect the way that genes are expressed and so, affect how a cell works. There is strong evidence that DNA methylation plays an important role in a range of disease. These processes can be highly dynamic, and are likely to mediate interactions between a cell’s innate genetic code and external factors in the environment. However, the complex relationship between DNA methylation, gene expression and disease is poorly understood. Recent technological advances mean that it is now possible to identify patterns of DNA methylation across the genome and identify genes that are active and repressed. These events do not occur in isolation – in fact there are highly coordinated networks of gene expression that mediate cellular phenotype. If we are to fully understand how these genomic processes mediate phenotype, we need an integrated view of the whole system. Here, we aim to combine several layers of genomic data to study a human neuronal cell treated with an antipsychotic drug. Very little is currently known about how these drugs work, and understanding the networks of genes affected by exposure may tell us more about the mechanisms involved in diseases such as schizophrenia. We will examine how DNA methylation affects gene expression with the goal of understanding the functional pathways induced by these medications. More broadly, this project will further our general understanding about the relationship between DNA methylation and gene expression and how these processes interact at a systems level.

EEG (a technique that measures the brain’s electrical signals) is a valuable aid to the diagnosis of epilepsy, but is relatively insensitive. A single routine EEG yields a clear positive result in only about 1/3 of patients with epilepsy. More sensitive approaches to the analysis of EEG would therefore be valuable. In this project we will explore whether mathematical analysis of EEG can enhance the diagnosis of a recently described form of temporal lobe epilepsy, transient epileptic amnesia (TEA). Patients with TEA present with recurring episodes of memory loss which are thought to have an epileptic basis: in some patients the distinction of TEA from other possible causes of transient amnesia is difficult. We will use advanced mathematical techniques to examine EEG from patients with temporal lobe epilepsy with clear epileptic abnormalities; patients with transient epileptic amnesia with clear epileptic abnormalities and patients with transient epileptic amnesia without clear epileptic abnormalities. If this work is successful it will allow earlier and more confident diagnosis of transient epileptic amnesia and contribute to the understanding of the underlying basis of this form of epilepsy.