Postgrad LIVE! Study Fairs

Birmingham | Edinburgh | Liverpool | Sheffield | Southampton | Bristol

University of Manchester Featured PhD Programmes
University of St Andrews Featured PhD Programmes
Coventry University Featured PhD Programmes
Bournemouth University Featured PhD Programmes
University of Reading Featured PhD Programmes

MRC DiMeN Doctoral Training Partnership: Accelerating super-resolution microscopy with machine learning and pattern recognition

This project is no longer listed in the FindAPhD
database and may not be available.

Click here to search the FindAPhD database
for PhD studentship opportunities
  • Full or part time
    Dr I Jayasinghe
    Prof N Gamper
    Dr J Leng
  • Application Deadline
    No more applications being accepted
  • Competition Funded PhD Project (European/UK Students Only)
    Competition Funded PhD Project (European/UK Students Only)

Project Description

Single molecule localisation microscopy (SMLM) is an invaluable tool for visualising molecular scale interactions and structures which are fundamental to life at the sub-cellular level. Super resolution images are assembled laboriously in SMLM experiments molecule by molecule; therefore it is a time intensive imaging modality which requires up to 90 mins to generate each image. These constraints have been a major barrier to its utility in the Medical Sciences. For example, the slow acquisition speed has made it undesirable for realtime live cell imaging experiments. In the proposed project, we will build a stand-alone programme which can reconstruct the fine spatial features of super-resolution images in a mere fraction (~ 10%) of the time needed for recording a full image with a standard protocol. The programme will achieve this by combining existing single molecule localisation algorithms implemented in Python and pattern-determination tools which we have developed to predict the final image iteratively. The software will use an iterative error estimation and image correction algorithm which will feed into a machine learning (ML) platform built-in with Python to inform the process of pattern determination (e.g. to identify structures which may be punctate, circular or linear). As the software learns iteratively, to perform pattern determination faster and more robustly, the time required to reconstruct the image using a live stream of primary image data is expected to be abbreviated significantly. This unprecedented level of speed will unlock a wide range of innovative and complex experiments as well as novel technologies which have not been possible thus far. For example, the added speed will enable molecular-scale mapping of rapid cellular events (in the order of seconds) in living disease models.

The programme of research will involve three principal objectives: (i) Integration of existing SMLM algorithms and pattern-determination models with a basic ML platform, (ii) application of the software to existing SMLM datasets of diverse cell types (neuronal cells, human cardiomyocytes and epithelial cells) and cell structures (calcium nanodomains, cytoskeleton, adhesion sites), and (iii) testing the algorithm for live image reconstruction, in conjunction with SMLM image acquisition. In the latter, we will characterise the accuracy of the software in rapidly reconstructing test structures as well as immunofluorescence labelling of different cell types. The successful candidate will work with Dr Isuru Jayasinghe (www.musclesuperres.com; Twitter @i_jayas), Prof Nikita Gamper (https://www.fbs.leeds.ac.uk/staff/profile.php?tag=Gamper) and Dr Joanna Leng (http://www.joannaleng.com/) who have combined expertise in super-resolution microscopy, cell biology, image analysis, visualisation and ML.

The outcome of the project will be a stand-alone programme which can be implemented in parallel with live SMLM experiments to visualise the super-resolution image faster and earlier in the image acquisition. The added speed will catalyse more widespread uptake of SMLM by Medical Scientists seeking to visualise the molecular underpinnings of human diseases. For example, this will enable time lapse imaging to observe, for the first time, the nanoscale membrane remodelling in cardiomyocytes which underpins cardiac dysfunction in heart patients.

Funding Notes

This studentship is part of the MRC Discovery Medicine North (DiMeN) partnership and is funded for 3.5 years. Including the following financial support:
Tax-free maintenance grant at the national UK Research Council rate
Full payment of tuition fees at the standard UK/EU rate
Research training support grant (RTSG)
Travel allowance for attendance at UK and international meetings
Opportunity to apply for Flexible Funds for further training and development
Please carefully read eligibility requirements and how to apply on our website, then use the link on this page to submit an application: https://goo.gl/jvPe1N

References

1. Jayasinghe, I., Clowsley, A.H., Green, E., Harrison, C., Lutz, T.,Baddeley, D., di Michele, L., Soeller, C. True molecular scale visualisation of variable clustering properties of ryanodine receptors. Cell Reports. 2018; 22(2) 557-567

2. Lin, R., Clowsley, A.H., Jayasinghe, I.D., Baddeley, D., Soeller, C.. Algorithmic corrections for localization microscopy with sCMOS cameras-characterisation of a computationally efficient localization approach. Optics Express. 2017; 25 (10), 11701-11716

3. Baddeley, D., Jayasinghe, I.D., Lam, L., Rossberger, S., Cannell, M.B., Soeller, C. Optical single-channel resolution imaging of the ryanodine receptor distribution in rat cardiac myocytes. Proc Natl Acad Sci U S A. 2009; 106 (52): 22275-80.



FindAPhD. Copyright 2005-2018
All rights reserved.