Developing time-gated Fourier Transform Raman micro-spectroscopy and data analysis for the cell sorter - Identification and sorting of cells by their chemical fingerprint

EPSRC funded DTP PhD Studentship within the Interdisciplinary Doctoral Training Hub “Physics of Life”

The ability of classifying cells has important implications in many branches of biology. Microscopic images are used to provide morphology patterns, such as nuclear-to-cytoplasmic ratio, granular cytoplasm, etc., which can be using machine learning approaches to improve pattern recognition and achieve unbiased cell classification. Beyond morphology, the features of the cells to be recognised can be enhanced by labelling molecules of interests and observing them by fluorescence microscopy. However, the additional fluorescing molecules introduced in the cells can alter their behaviour and complicate preparation. Among different label-free alternatives, Raman scattering spectroscopy has the advantage to provide a chemical fingerprint of the specimen under investigation. In this technique, the spectrum of the light scattered by the sample upon laser excitation is collected and the chemical composition can be retrieved by analysing the frequency shifts given by vibrational frequencies of the constituting molecules. Hooke Instruments (partner in this project) is developing a new Raman detection instrument based on time-correlated Fourier Transform spectroscopy. The aim is to separate the Raman signal from the cell autofluorescence using their different time dependence, to improve the instrument sensitivity. The instrument will be equipped with a pulsed laser and time-correlated detection for time resolved measurements.

The aim of this PhD project is to develop algorithms to extract the relevant quantities and feed-back to the instrument operation. Further, to extract the spectro-temporal signatures and analyse them for identification and classification purposes. By combining multivariate and machine learning techniques [1], the student will develop analysis tools to identify and sort cells and bacteria with high accuracy using their chemical signatures. The project is then envisaged to apply and refine the data analysis techniques to increasingly challenging and important problems in cell sorting, and Raman imaging.

Research Environment:

This project is the result of a long-standing collaboration between the Lead and Joint-Lead supervisors and Hooke Instruments Ltd. The student will be part of the vibrant multi-disciplinary research environment of the Quantum Optoelectronics and Biophotonics group. The group research activities span across a range of topics, from solid state physics to bio-imaging and biosensing and image / big data analysis. The student will join regular meetings where group members will discuss their research and he/she will be encouraged to attend to weekly seminars at the School of Physics and School of Biosciences.

Training and Development Opportunities:

The student will gain a range of experimental and programming research skills. The student will learn to program in Matlab, a software used in many applications, within academia and industry. The multivariate and machine learning based data analysis techniques that the student will apply are highly transferable skills useful to employers across various jobs and industries. Beyond data analysis, the student will be trained in Raman micro-spectroscopy, and will be involved in the development of the software to interface the new instrument. Thanks to the involvement of the company, the student will experience the process of product development in a commercial R&D setting. The student will disseminate the results of the project through publications and participation at national and international conferences, developing writing and presentation skills.

 Co-Supervisor for this project is Dr Bei Li (Hooke Instruments Ltd)

Deadline for applications

3rd July 2022. We may however close this opportunity earlier if a suitable candidate is identified.