2010 intake

   
Thomas Bartlett Network models of stochastic processes relating to DNA methylation and stem cell and cancer biology
Sian Culley Super Resolution Fluorescence Microscopy
Ryan Dee Development of a small diameter tissue engineered arterial graft using autologous stem cells for a paediatric application
Andrea Dimitracopoulos Role of microtubule-cortex interactions in mitotic spindle positioning
Michael Epstein Differential Geometric Methods for Bayesian Inference in Ion Channel Models
Tharindi Hapuarachchi Shedding light into brain energy metabolism in neonatal hypoxic-ischaemia using novel multimodal brain imaging measurements and in silico models
Sonja Lehtinen Exploring Cellular Ageing and Stress Response Through Network Analysis
Vivien Li The role of Connexin-43 in the pathogenesis of liver disease: A systems approach
Nicholas Mcgranahan Bioinformatics-driven Identification of Functional Regulators of Cancer Genome Instability
Fintan Nagle Geometrically modelling the brain's representation of human faces
Emmanouil Protonotarios Local Symmetry Analysis and Psychophysics of visual textures and spatial patterns
Tom Roberts High Resolution 3D µMRI for Phenotyping & Cardiac Imaging
Robert Stanley Modelling and predicting the behaviour of the Phospholipase D, PI(4)P5Kinase, Arf G-protein signalling triplet
Charlotte Strandkvist The role of stochastic processes in development
Janine Symonds Models of cellular purine nucleotide metabolism as predictive tools in health and disease
   
   

Thomas Bartlett

Network models of stochastic processes relating to DNA methylation and stem cell and cancer biology

Supervisors: Prof. Alexey Zaikin, Prof. Sofia Olhede and Prof. Martin Widschwendter


Complex systems which can be modelled as networks are ubiquitous. Well-known examples include social and economic networks, as well as many examples in cell biology such as gene regulatory and protein signalling networks. Many cell biological processes are inherently stochastic and non-stationary, and this is the perspective from which we are developing novel statistical network models. These models are motivated by cell biological processes relating to DNA methylation and stem cell and cancer biology, and can be generalised to other systems.

bartlettr

Sian Culley


Super Resolution Fluorescence Microscopy


Supervisors: Dr. Angus Bain, Prof. Jonathan Ashmore

Although fluorescence microscopy is a common tool in biological research, conventional microscopes are restricted to resolutions of approximately half the wavelength of illuminating light. Many biological structures of interest, such as individual proteins and vesicles, therefore cannot be resolved. Recent advances using stimulated emission depletion (STED), in the technique of STED microscopy, have broken this resolution limit. In this research, STED microscope systems will be developed with a view to making super resolution fluorescence microscopy available at UCL. In particular, one aim is to investigate the physiology of inner hair cells using dyes such as synaptopHluorin which is targeted to synaptic vesicles.

culleyr

Ryan Dee

Development of a small diameter tissue engineered arterial graft using autologous stem cells for a paediatric application

Supervisors: Prof. Alex Seifalian (Centre for Nanotechnology and Regenerative Medicine), Dr. Paolo De Coppi (Institute of Child Health)

Many congenital conditions especially congenital heart defects require the use of a bypass graft to reduce morbidity. Suitable autologous grafts may not always be available and so synthetic alternatives must be found. In adults a non-biodegradable biomaterials approach can be used. If such an approach were used in paediatric patients then further surgical intervention would be required to replace the graft as the patient grows. What is required is a graft that can grow somatically with the patient. This can be achieved using tissue engineering (TE) and autologous stem cells.

The project will involve the development of a small diameter graft using a novel biodegradable nanocomposite developed at UCL, POSS-PCL. We will look at optimising the mechanical, structural and biodegradable properties of the material for this application. We will then look at sources for autologous stem cells, techniques for differentiation and ways of further biofunctionalising our TE scaffold material.

dee

Andrea Dimitracopoulos

Role of microtubule-cortex interactions in mitotic spindle positioning

Supervisors: Prof. Buzz Baum and Prof. Tom Duke

Abstract: Proper assembly, positioning and orientation of the mitotic spindle is critical for tissue morphogenesis in many systems. The microtubules apparently read marks in the cell cortex that mirror asymmetric cues in the extracellular space, thereby obtaining information about the dividing cell’s immediate environment. It remains unclear, however, exactly how the mechanical coupling between astral microtubules and the actin cortex enables the spindle to find its correct position and orientation. This project aims to use a combination of mathematical modelling and experimentation on model systems to elucidate the molecular basis of this process.

dimitracopoulos
Michael Epstein

Differential Geometric Methods for Bayesian Inference in Ion Channel Models


Supervisors: Prof. Mark Girolami (Statistical Science) and Prof. Lucia Sivilotti (Pharmacology)

Abstract: Ion channels are transmembrane proteins present in cells which are crucial for a range of important physiological processes, such as modulating the synaptic connection between neurones, or between nerves and muscle. The conformational changes which describe how ion channels open and close are statistically well modelled as aggregated Markov processes. Postulated Markov models are fitted from single channel recordings using different agonists and agonist concentrations. However, it is becoming increasingly important to discriminate between different plausible Markov models fitted from these data.

This project therefore seeks to employ recent advances made in differential geometric MCMC methods to perform Bayesian inference over the fitted models. This will allow rational and systematic analysis of competing Markov models. This approach, allied to structural data from X-ray crystallography, aims to provide a better understanding of the relation between structure and function of ion-channels. It is expected that this will help inform the process of rational drug design for specific ion channel targets.

epstein

Tharindi Hapuarachchi

Shedding light into brain energy metabolism in neonatal hypoxic-ischaemia using novel multimodal brain imaging measurements and in silico models

Supervisors: Dr. Ilias Tachtsidis (Department of Medical Physics and Bioengineering) and Dr. Nicola Robertson (Institute for Women's Health)

Neonatal cerebral hypoxic-ischemia is a condition resulting from birth asphyxia - reduced oxygen delivery or/and blood flow, occurring either in utero or during delivery. My PhD project will involve the development/expansion of a computational model of circulation and metabolism in the neonatal brain. It incorporates both Near-Infrared Spectroscopy (NIRS) and Magnetic Resonance Spectroscopy (MRS) measurements. Results from ongoing experiments involving brief anoxia in piglets will be used to help test and develop the model. 

We hope that this work will lead to the evolution of a new tool that integrates multimodal measurements of brain tissue with in silico modelling – allowing us to test different hypotheses of brain energetics and interpret experimental data towards information of clinical importance.

hapuarachchi
Sonja Lehtinen

Exploring Cellular Ageing and Stress Response Through Network Analysis


Supervisors: Prof. Christine Orengo and Prof. Jürg Bähler.

Target of Rapamycin (TOR) is a central hub in a signalling network coordinating cellular response to nutrient availability, growth factors and environmental stress. The pathway is of great interest because TOR signalling has been linked with ageing and age-related disease in organisms ranging from yeast to human.

The aim of my project is to extend characterization of the TOR pathway. A variety of existing methods, both experimental and computational, yield information about physical interaction and functional association between proteins. I will be exploring statistical methods for combining these protein-protein networks and  using them to predict new TOR pathway components. These predictions will then be tested experimentally and the results used to further refine the prediction methods.

lehtinen

Vivien Li


The role of Connexin-43 in the pathogenesis of liver disease: A systems approach.

Supervisors: Prof. David Bogle (Chemical Engineering), Prof. Rajiv Jalan (Hepatology)

Liver disease is the 5th most common cause of death in the UK. Yet unlike other leading causes of death, liver disease rates are increasing rather than declining; particularly in Non Alcoholic Fatty Liver Disease (NAFLD). It affects up to 25% of the US and UK population respectively, and even in the countries that are previously exempt, such as the Far East and China, are gradually being affected by the adoptions of the western lifestyles. Alarmingly, it tends to progress to steatohepatitis (NASH), fibrosis, cirrhosis and liver cancer. Current strategies for the management of these disorders have mainly been based on intracellular signalling pathways and have ignored the role of alterations in cell-cell communication in the development of the disease and as a potential therapeutic target in the prevention of progression and or the treatment of these conditions.

The main aim of this project is to test the hypotheses that liver disease is characterized by a shift in expression of the Connexin proteins (which control direct cell–cell communication) from Cx26 to Cx43; and that the modulation of the increased expression of Cx43 would prevent progression of NALFD, liver injury, fibrosis and cirrhosis in appropriate models of liver disease. These hypotheses will be tested through histological and animal experiments. The data generated will be aimed to develop an in silico multicellular systems model of hepatic cell-cell communication that can be used as a tool to improve understanding both of normal behaviour of the liver and the functional consequences of liver disease, as well as the potential to develop therapeutic interventions targeting Cx43. This project may also integrate this in silico systems model of hepatocytes into the existing UCL beacon project, which is a ‘composite’ computational model consisting of nine- linked simple models that simulate the interactive behaviour between liver, pancreas and blood in glucose homeostasis. This will allow insights to the role of hepatocyte-hepatocyte communication in the progression from insulin resistance to type 2 diabetes, and the corresponding diabetic effects in promoting NAFLD and beyond.

li

Nicholas Mcgranahan


Bioinformatics-driven Identification of Functional Regulators of Cancer Genome Instability


Supervisors: Dr. Charles Swanton and Dr. Andrew Teschendorff

Genomic instability is a characteristic of the majority of human cancers. The most common form of genomic instability, chromosomal instability (CIN) is associated with intra-tumour heterogeneity, drug resistance, and poor patient outcome. Despite the prevalence and clinical importance of CIN, a molecular basis underpinning its initiation and development is poorly understood.

New technologies, such as next-generation sequencing (NGS), have led to an avalanche of genomic cancer data. Using these data in meaningful ways will be crucial to aid our understanding of cancer, and, in particular genomic instability. My PhD project will aim to integrate data from a large number of sources, including human NGS tumour genomes, mRNA expression, single nucleotide polymorphism (SNP), genomic copy number, RNAi screening, drug resistance and survival data. The data will be used to explore patterns of genomic instability and to facilitate the identification of function regulators of intra-tumour heterogeneity and chromosomal instability. The analysis will make use of the integrative machine-learning approach.

mcgranahan

Fintan Nagle

Geometrically modelling the brain's representation of human faces

supervisors:Prof. Alan Johnston (Cognitive, Perceptual and Brain Sciences, UCL) and Prof. Peter McOwan (Computer Science, Queen Mary)

The human brain is highly specialised at recognising and detecting faces, their expressions, and their identity. We know that the fusiform face area of the brain is where much of this processing takes places, and we have very basic conceptual models of how faces are encoded and recognised. I am seeking a more formalised, mathematical model, including 3D concepts of shape using the methods of differential geometry.


nagler

Emmanouil Protonotarios

Local Symmetry Analysis and Psychophysics of visual textures and spatial patterns


Supervisors: Dr. Lewis Griffin and Prof. Alan Johnston

Human vision relying on complicated neural processes performs significantly better in vision related tasks compared to artificially developed algorithms. The study of the mechanisms that human vision incorporates for this amazing performance can benefit the development of more effective methods for encoding visual information.

My PhD will include:
1. Analysis and characterization of visual textures and spatial patterns based on methods such as Basic Image Features (BIFs) which have been derived by consideration of the analysis conducted by the visual system on the local symmetry of fragments of the image as well as development of alternative such methods.
2. Psychophysical analysis of the human performance in tasks such as texture discrimination in order to understand how the human brain represents visual information and to improve texture encoding algorithms.

protonotarios
Tom Roberts

High Resolution 3D µMRI for Phenotyping & Cardiac Imaging


Supervisors: Dr. Mark Lythgoe (Centre for Advanced Biomedical Imaging) & Prof. Peter Scambler (Institute of Child Health)

Understanding the functions of the human genome is one of the most exciting and promising areas for disease diagnostics and treatment, particularly in cardiac disease which is an increasing health burden.

To aid our understanding of human diseases, large international programs are creating thousands of new mouse models with mutations across the genome. These transgenic mice may have a genetically-induced heart defect, for instance. Studying such mouse models will give us new insights into the disease and treatment therapies. However, examining individual mouse models is both time-consuming and very costly.

The field of phenotyping has emerged with the aim of characterising the phenotypes associated with each mutation in the genome. High resolution MRI is emerging as one possible technique for rapidly identifying phenotypes in mouse models. My PhD will contribute to improving the phenotyping process with a particular focus on cardiac diseases. This will involve the development of µMRI techniques, enhancing MRI sensitivity and working with new MR sequences and contrast agents.

Tom's Homepage: http://www.ucl.ac.uk/~ucbptar/


robertsr

Robert Stanley

Modelling and predicting the behaviour of the Phospholipase D, PI(4)P5Kinase, Arf G-protein signalling triplet


Supervisors: Prof. Geraint Thomas & Dr. Kevin Bryson

The intracellular signalling network comprising of the activation of the enzymes phospholipase D (PLD) and phosphatidylinositol 4-phosphate 5-kinase (PIP5K) by the ARF family of proteins has a novel 'cross-talk' structure in that the products of PLD and PIP5K have a role in the activation of the other enzyme. These interactions are thought to control processes such as actin cytoskeletal dynamics, membrane trafficking, and exocytosis. Within the cardiovascular system this signalling pathway has been implicated in cardiovascular development, vascular stability, and angiogenesis.

The aim of my PhD is to build and refine mathematical models of this system, and determine the robustness and biological relevance of behaviours that are observed in these models. These results will subsequently contribute to the design of experiments to determine the veracity of my models and the particular characteristics of the biological signalling network.

stanley

Charlotte Strandkvist

The role of stochastic processes in development

Supervisors: Prof. Buzz Baum (MRC LMCB), Dr. Alexandre Kabla (University of Cambridge), Prof. Johan Paulsson (Harvard Systems Biology)

Biological systems are inherently noisy. Understanding how embryonic development emerges in a robust deterministic manner is one of the central challenges of modern biology. Genetically identical cells exhibit remarkable compositional and behavioural heterogeneity. Recent work from the Paulsson lab has demonstrated that randomness in the partitioning of proteins and organelles at cell division is an important contributor of noise in the system and may account for much of the cell-to-cell variability that has been attributed to fluctuations in gene expression.

The aim of this project  is to develop a theoretical framework to address how stochasticity in the partitioning of organelles at cell division contributes to heterogeneity at the population level. Our focus will be the segregation of mitochondria. Using mathematical analysis and computational modelling we will attempt to connect the available data to falsifiable hypotheses about mitochondrial dynamics and identify underlying principles that are conserved across species.

Charlotte Strandkvist

Janine Symonds

Models of cellular purine nucleotide metabolism as predictive tools in health and disease

Supervisors: Dr. Kevin Bryson (Computer Science) and Prof. Geraint Thomas (Cell & Developmental Biology)

Purines are crucial cellular components, functioning both as the building blocks of nucleic acids and as intermediate metabolites such as the ubiquitous ATP. Purine metabolism, which is responsible for the synthesis, breakdown and inter-conversion of purines, is a complex network involving multiple feedback mechanisms. This network is the target for various therapeutics which are used in the treatment of a wide range of diseases including cancer and immune disorders. Although a series of general mathematical models of human purine metabolism have been produced they may not be optimal for certain tissues and diseases as different tissues may have different purine requirements.

My project aims to establish new models of purine metabolism tailored to specific cell types or disease states which could then be used to model the effects of pharmaceuticals on purine metabolism.

symonds

Page last modified on 22 jan 14 21:15