Supervisors:  Sylwia Bujkiewicz and Michael Sweeting (University of Leicester)

Recent advances have led to the discovery of a multitude of pharmacological therapies in cancer. Many of them are targeted to small subsets of the population, for example those patients who are positive for a particular genetic biomarker. When clinical trials of targeted therapies are based on small samples of patients, their effectiveness may be obtained with large uncertainty leading to challenges when licencing decisions about the market access of the new health technologies need to be made. This project brings together a range of artificial intelligence methods to inform a complex decision-making process at the licencing and reimbursement stages by the regulatory and health technology assessment agencies. A complex decision-modelling framework informed by effectiveness parameters obtained using advanced Bayesian evidence synthesis techniques will be developed. Information will be combined from diverse sources of evidence, including surrogate endpoints, and applied to a case study in advanced colorectal cancer.

Supervisors:  Andrew French, Maura Corsetti, Yorgos Tzimiropoulos, Michael Pound, Penny Gowland and Caroline Hoad (University of Nottingham)

Machine learning is changing how medical images are analysed. This project will develop new deep learning approaches using a library of analysed images from multimodal MRI datasets of the gastrointestinal tract. The student will develop new algorithmic approaches to deep learning, to automate analysis and improve accuracy of interpretation.

Alterations in GI motility, secretion and transit impact on diseases such as irritable bowel syndrome and inflammatory bowel disease. Until recently it was not possible to evaluate different aspects of GI function in a single examination, but this is now possible, applying novel non-invasive MR measures developed at Nottingham. However this requires very labour intensive analysis (eg manually outlining the bowel wall in 3D data) and is particular to certain MRI machines, hindering the porting of these methods to different scanners, sites, field strengths, etc. These issues are significant barriers to translation from experimental to clinical medicine.Accordion content

Academic supervisors: Georgious Gkoutos (University of Birmingham), Richard Emes (University of Nottingham) and Philip Quinlan (University of Nottingham)

Correctly linking patient information to tissue samples allows biomedical discovery to benefit patients and drive research. However this requires extensive joining of data and hands-on conversion of data types. Because of this lack of clinical annotation, It is known that in the UK there are millions of samples that remain unused.

The UKCRC Tissue Directory and Coordination Centre at Nottingham University aims to overcome these difficulties by allowing researchers to find the biobank that is most likely to have the samples required to support their research. The aim of this project is to work with UKCRC TDCC and a commercial partner (BC Platforms) to research semantic interoperability in terms of how biomedical data can be better represented.

We will utilise the artificial intelligence technologies in the fields of text mining, semantic interoperability, as well as semantic AI so to develop a health research information environment amenable to data mining.

Supervisors:  Jonathan Hirst and Thomas Gaertner (University of Nottingham)

In this PhD project, we will extend state-of-the-art artificial intelligence algorithms to help discover new drug molecules. The springboard for the research is a unique collaboration between Prof Jonathan Hirst (Chemistry), Prof Thomas Gaertner (Computer Science) at the University of Nottingham and the pharmaceutical company GlaxoSmithKline (GSK), which is synthesizing and testing new molecules for effectiveness as a potential treatment for a complex and chronic lung disease, idiopathic pulmonary fibrosis. We will develop new AI methods, focused on the specific demands of drug discovery. We will assess approaches such as long short-term memory, variational autoencoders, convolutional neural networks and Bayesian kernel methods. Starting from a firm theoretical foundation, we expect to derive algorithms that are more successful and more generally applicable than earlier work. Furthermore, we will evaluate the models performance by selecting a subset of predicted structures, synthesising them and testing them in established assays at GSK.

Supervisors:  Kamlesh Khunti, Clare Gillies, Francesco Zaccardi and Laura Gray (University of Leicester)

In making treatment decisions clinicians must take into account evidence from relevant randomised controlled trials (RCTs). Unfortunately the results of RCTs will never be relevant to all patients and settings, and hence the effectiveness of clinical interventions using real world evidence is becoming increasingly important. For this proposal we plan to evaluate the effectiveness of SGLT-2 and GLP-1RAs for type 2 diabetes, comparing results from RCTs with real world evidence from the Clinical Practice Research Datalink (CPRD). We will:

i)  Extract patient data from CPRD to assess if RCT results from relevant trials are replicated in a matched UK population.

ii)  Assess effectiveness of SGLT-2 inhibitors and GLP-1RAs in the clinical population who receive the treatment in the UK (which will be more diverse than the RCT populations).

Supervisors:  Dmitry Veprintsev (University of Nottingham) and Iain Styles (University of Birmingham)

This project will pave the way for Artificial Intelligence-based data-driven “engineering” of more efficacious drugs with fewer side-effects. Around one-third of drugs target G protein coupled receptors (GPCRs), but they typically activate multiple signalling pathways, reducing the efficacy and increasing undesirable side effects.

Designing better drugs requires a more detailed understanding of their mechanisms of action.  We will develop state-of-the art AI and machine learning approaches to combine unique pathway activation data from alanine scanning with structural information and high-throughput screens of molecular compound libraries to understand the molecular basis of drug action in GPCRs, predict the action of a large range of drugs across GPCR classes, and predict ligands that will give rise to a desired signalling profile. Applicants should have a first degree in computer science (or similar), interests in machine learning, and enthusiasm for interdisciplinary work with the supervisors, Prof Veprintsev and Dr Styles.

Supervisors:  Christopher Yau, Andrew Beggs (University of Birmingham) and Theodore Kypraios (University of Nottingham)

This exciting PhD project will use Artificial Intelligence techniques to construct a state-of-the-art computer simulation model of an organoid cancer model or “mini-tumour”. This computer model will enable cancer biologists to conduct in-silico experiments to predict the likely response of a mini-tumour to a novel treatment before it is confirmed with an actual experiment. Importantly, in the construction of this model, we will also learn more about how cancers function. These models will be trained using high-dimensional experimental data that defy human interpretation and require a machine-based learning system to extract important relationships. Building this computer model will help us to learn the “rules” which govern the behaviour of these mini-tumours and how cancers evolve. This project will involve an exciting supervisory team spanning Birmingham, Nottingham and the Alan Turing Institute – the National Centre for Artificial Intelligence and Data Science. For more information, see Yau group website (http://cwcyau.github.io).

Supervisors:  Guruprasad Aithal, Andrew Bennett, Simon Macdonald and Jane Grove

In the liver, several av-containing integrin heterodimers have been described: aVb1, aVb3, aVb5, aVb6 and aVb8. Integrins are involved in the activation of TGFb and mechanosensing of tissue stiffness. We will 1) determine the pattern and level of expression of av and b1,3,5,6 and 8 integrins at the protein level in human liver samples from patients with varying degrees of fibrosis and use co-localization with activated hepatic stellate cell markers to indicate potential involvement with disease, 2) establish a primary human stellate culture system on hydrogels of varying stiffness to model fibrosis in vitro and determine expression of av-containing integrin heterodimers at varying stages of stellate cell activation, 3) assess efficacy of a panel of soluble integrin inhibitors upon stellate cell activation examining TGFb activation and stiffness induced signal transduction and 4) determine whether soluble integrin inhibitors can reverse stellate cell activation to a quiescent state once fibrogenic processes are established.

Supervisors:  Kenton Arkill, Dave Bates and Cathy Merry

When fluid and proteins leak out of the blood into the tissues, severe disease can occur, and this happens acutely in sepsis and cancer, but chronically in diabetes.. Capillary walls, including the endothelial glycocalyx, control molecular movement out of the vasculature. This glycocalyx is a size and charge filter that is disrupted in pathological states, and a key consideration in drug delivery methods and fluid therapies. The filter structure remains poorly defined due to incomplete in-vitro models.

This project will use new cryo-electron microscopy and ion beam mass spectroscopy techniques to define the physiological structure, and the mechanisms behind glycocalyx control. The outcomes will form the bedrock to elucidate how pathologies such as diabetes, can be therapeutically rectified.

The student will join Dr Arkill’s multidisciplinary team based in the Tumour and Vascular Research Laboratories but includes training and use of Glycobiology and the Nanoscale and Microscale Research Centre. 

Supervisors:  Richard Bowtell, Dorothee Auer, Penny Gowland and Galina Pavlovskaya

The mechanisms of brain iron changes underlying neurodegenerative diseases are largely unknown. One hypothesis suggests that erythrocytes leak through an impaired blood-brain barrier leading to activation of microglia. This results in intracellular deposition of haemosiderin, a disorganised iron storage complex which contains unbound iron ions. In this state, iron is neurotoxic producing free radicals and causing oxidative stress. The nucleus basalis of Meynert is a cholinergic basal forebrain nucleus which is affected early in the course of many neuropsychiatric disorders. Brain iron can be detected using gradient-echo MRI with areas of high iron appearing hypo-intense in magnitude images. Advanced susceptibility mapping at high field is needed for reliable quantification. Disentangling the mechanisms that lead to iron-mediated neurotoxicity is at the frontier of multidisciplinary research and clinical imaging. Moreover, non-invasive iron mapping using MRI provides a mechanistic biomarker for disease prediction that can be exploited in future clinical trials.

Supervisors:  Galina Pavlovskaya, Alan Knox, Dominick Shaw and Richard Graham

Airway mucus plugs cause airway obstruction in patients with airways disease, such as cystic fibrosis and bronchiectasis. Little is known about the effects of therapy on plug composition, airway ventilation or clinical outcome. Mucin (a major component of the plug) degradation is critical for plug clearance but in vivo methods monitoring this are lacking. We hypothesise that these changes in plug physical properties can be captured by ²³Na MRI methods and propose to apply multi-scale ²³Na MRI for this purpose. We will use the 9.4T scanner to tune up the relevant ²³Na contrast at the molecular level ex-vivo using sputa samples of healthy volunteers and patients with bronchiectasis and cystic fibrosis pre- and post-intervention (carbocysteine/nebulised DNAse/ivacaftor) for novel ²³Na whole body lung imaging at 3T and 7T in health and disease at SPMIC Nottingham. We complement results of these findings with mathematical modelling.

Supervisors:  Dmitry Veprintsev, Stephen Briddon and Mark Wheatley (University of Birmingham)|

G protein coupled receptors convert extracellular signals mediated by hormones and neurotransmitters into intracellular responses. Interactions between receptors could lead to dramatic changes in their signalling properties and significantly expand the “pharmacological universe” of GPCRs. The aim of the project is to understand the molecular basis for functional modulation of receptors in hetero-dimers. We will combine alanine scanning mutagenesis with the state of the art FRET and BRET-based techniques to study the impact of mutations on dimerisation and signalling properties of CB2 and CCR5. While the large datasets generated will be too complex for manual analysis, the quantity of data does however allow modern machine learning techniques to be utilised. We will develop novel machine-learning approaches that can  model the complex relationships between the many interdependent variables, and to incorporate existing knowledge of the structure and function of GPCRs to augment and condition the learning. 

 

Supervisors:  Martine Hamann and Flaviano Giorgini

Studies on noise-induced hearing loss classically focus on the drastic damage inflicted upon cochlear hair cells. Noise-induced hearing loss is often visible after a few months, as the cochlear damage becomes irreversible. Stress granules are cytoplasmic aggregates of proteins and RNAs that appear under stress conditions. Our recent studies have identified stress granules in cochlear hair cells minutes following acoustic over-exposure. The project will explore whether cochlear stress granules are biomarkers of the early stages of hidden hearing loss. The presence of stress granules will be evaluated by immunofluorescence for specific stress granule markers.  Proteomic and transcriptomic analysis will provide mechanistic insights into their function and their specific link to signalling pathways. The project will also test the ability of antioxidants to reduce the presence of stress granules and stop or delay the damage of cochlear structures that is indicative of permanent hearing loss.

Supervisors:  Toru Suzuki, Marco Oggioni and Liam Heaney

This multidisciplinary project combines the use of microbiology, in vivo research models and analytical science to identify bacterial contributions for the production of trimethylamine (TMA). TMAO, a downstream metabolite of TMA, is a circulating metabolic biomarker of the gut microbiome that has been shown to associate with increased severity and adverse events (e.g. mortality/hospitalisation) in cardiovascular diseases including heart failure and myocardial infarction. A series of projects will be performed to identify the bacteria responsible for the production of TMA in the gut, as well as those capable of metabolising TMA as a potential therapeutic intervention to reduce the bioavailability of TMA and therefore reduce circulating levels of TMAO. The projects will include the use of next generation sequencing for microbiome analysis and gene expression profiling, animal models of bacterial infection, and bio fluid analyses using common laboratory techniques and advanced techniques including liquid chromatography-mass spectrometry for metabolome analysis.

Supervisors:  Louise Wain (University of Leicester) and Gisli Jenkins (University of Nottingham)

Idiopathic Pulmonary Fibrosis (IPF) is a rare, chronic and progressive lung disease with poor prognosis and limited treatment options. More men than women are diagnosed with IPF and the primary aim of this project will be to understand the genetic factors that drive this difference in order to further our understanding of the disease process to aid development of new therapeutic strategies.  The student will be based within an internationally-recognised respiratory genetic epidemiology group and benefit from close collaborations with leading IPF clinicians and researchers from the UK and USA.  A broad training in genetic epidemiology will equip the student for a career in the fast-moving and opportunity-rich field of human disease genetics. Applicants should have a background in bioinformatics, epidemiology, statistics or genetics (or similar), with an aptitude for computing (prior programming experience advantageous but not essential) and a keen interest in how genetics affects human health and disease.