Developing vision and learning methods to derive physical models from microscopy
Supervisors: Iain Styles, Ales Leonardis, Fabian Spill, Rob Neely, Steve Thomas, Steve Watson
This exciting interdisciplinary project,supported by a team of supervisors with complementary expertise in computer science, imaging, mathematical modelling and biology, will develop new computer vision and machine learning methods to understand the dynamics of the cellular cytoskelecton from microscopic images, and to explain the dynamics using physical models. The project is suitable for students with a strong background in Computer Science, Physics, or Mathematics who would like to apply their skills to develop fundamentally new methods to understand an important problem in biology.
The cytoskeleton (a complex network of proteins) in cells plays a crucial role in their dynamics and function. It maintains the cell’s structure and is implicated in the formation of various structures including lamellipodia, stress fibres and podosomes, which serve to enable cell motility. Dysfunction of the cytoskeleton’s dynamics is known to impede thrombosis, an important wound healing process, and it has been suggested that addressing this could provide possible therapeutic pathways. There are also deep connections to cancer therapy, where drugs such as Paclitaxel work by inhibiting the dynamics of the cytoskeleton. Understanding its dynamics is therefore vital to fully understand these diseases and how they may be treated.
We will study the cytoskeleton using a range of advanced fluorescence microscopy techniques including TIRF, diSPIM and lattice light sheet techniques for live-cell super-resolution imaging and N-STORM, for ultra high-resolution single molecule imaging. We will use a combination of classical image processing and machine learning to segment the cytoskeleton and identify and categorise its sub-components. We will then map this information onto physical models of the cytoskeleton using an inference process that will allow key parameters of those models to be estimated and a quantitative understanding of the cytoskeleton’s dynamics developed.
The project will develop fundamentally new computational models and techniques that will in turn provide new insight into cellular and subcellular dynamic processes, and potentially suggest new mechanisms of disease that could be targeted by therapeutic agents. We expect that both the methods themselves, and the biological insights that they yield will generate high-quality publications. During the project, you will develop advanced skills in computer vision, image processing, machine learning, and mathematical modelling, and will also gain an understanding of the practical aspects of cellular and subcellular biology. To support you in gaining these skills, you will undertake a range of interdisciplinary taught and practical skills modules at the outset of the PhD.
In addition to the facilities and resources provided by the CDT, you will benefit from access to a wide range of state-of-the-art microscopy techniques in the £10m Centre of Membrane Proteins and Receptors. This includes the only lattice light sheet microscope in the UK. In addition to the imaging facilities, COMPARE has outstanding dedicated high-performance computing capabilities that will be available to you. There is a potential opportunity to spend time in the US laboratory of an overseas collaborator.
This PhD project sits within the EPSRC Centre for Doctoral Training in Physical Sciences for Health (Sci-Phy-4-Health) & encompasses a 4 year Integrated PhD programme. Each year the CDT is able to offer a fully funded scholarship from EPSRC to UK students (tax free stipend £14,553 for entrance in 2017/2018).
EU students are eligible for awards of tuition fees only but may also be eligible for stipend funding if they have studied or worked in the UK for at least 3 continuous years immediately prior to course entry. Candidates are advised to check the EPSRC website for eligibility details.
How good is research at University of Birmingham in Chemistry?
FTE Category A staff submitted: 28.00
Research output data provided by the Research Excellence Framework (REF)
Click here to see the results for all UK universities