Histopathological images are routinely used to characterize complex phenotypes. For example, pathologists regularly study stained images of tissue biopsies for cancer diagnosis as cancer is known to change the morphological features of cells including cell shape and size [1]. In recent years, computational approaches have been developed to extract morphological, textural and other relevant image features from histological images [2]. These image features are then used for the classification of the diseased tissue. In cancer tissues use of machine learning algorithms for image and genomic analysis have been quite successful [3].
However, all these methods have used bulk gene expression data to study this correlation. Bulk gene expression will only give an average of the expression within a tissue area and it will not be possible to understand how a drug's efficacy or toxicity will impact different cell types within a tissue region. By contrast single cell sequencing gives individual molecular measurement for each single cell in a tissue.
The main aim of this project is to develop a machine learning algorithm to identify sets of genes and the pathways they are involved in which are correlated with image features from histological images. The student will use Spatial Transcriptomics (ST) data that gives near single cell resolution gene expression data along with its spatial location [4]. ST maps whole transcriptome data with morphological context in tissue architecture. This will link the prognostic markers from the images with molecular markers. The student will develop a deep learning model that will extract relevant image features from H&E stained images in a latent space and use non-negative matrix factorization (NMF) and graph neural network to associate these features with gene expression. In addition to histopathological data we will focus on specific pathways that are known to influence liver fibrosis. As an example, matrix components that make up the scar (e.g. COL1A1), markers of myofibroblasts responsible for scar deposition (e.g. aSMA) and pro-fibrotic pathways and their cellular modulators such as TGFbeta and macrophage subsets.
The student will also develop an interactive user-friendly tool using python and Javascript. The pathologists will then give histopathology images as input and the tool will predict the possible pathways that are enriched with a certain region in the image.
Eligibility
Applicants must have obtained or be about to obtain a First or Upper Second class UK honours degree, or the equivalent qualifications gained outside the UK, in a relevant discipline.
Before you Apply
Applicants must make direct contact with preferred supervisors before applying. It is your responsibility to make arrangements to meet with potential supervisors, prior to submitting a formal online application.
How to Apply
To be considered for this project you MUST submit a formal online application form - full details on how to apply can be found on the MRC DTP website https://www.bmh.manchester.ac.uk/study/research/mrc-dtp/
Your application form must be accompanied by a number of supporting documents by the advertised deadlines. Without all the required documents submitted at the time of application, your application will not be processed and we cannot accept responsibility for late or missed deadlines. Incomplete applications will not be considered. If you have any queries regarding making an application please contact our admissions team [Email Address Removed]
Equality, Diversity and Inclusion
Equality, diversity and inclusion is fundamental to the success of The University of Manchester, and is at the heart of all of our activities. The full Equality, diversity and inclusion statement can be found on the website https://www.bmh.manchester.ac.uk/study/research/apply/equality-diversity-inclusion/