Projects available

The role of Proteomics in Accelerating Translational Insights for Drug Discovery and Development 

This project will focus on advancing our understanding of the circulating proteome. You will study how proteomics can be utilised as a tool for identifying tissue dysfunction underlying diverse diseases and identifying protein signatures associated with drug intake and response, using data from population-based cohort studies and trials. 

You will learn and carry out a range of analytical approaches to identify proteins whose plasma abundance is associated with use of each of the drugs and/or drug classes of interest.  You will also develop and use machine learning algorithms to prioritise molecular markers based on a range of weighting factors, specifically technical parameters.

The project is part of the AI for Drug Discovery programme, and will benefit from being embedded within the Centre for Multi-omics in the Precision Healthcare University Research Institute (PHURI). The Centre is one of the world leaders in the genetic discovery of molecular traits, such as metabolites and proteins, which we use to identify shared genetic regulation with common, complex diseases to discover new mechanistic insights. We have a keen interest in training the next generation of scientists that can translate data into actionable insights.

Eligibility and Applying 

This project is part of the UKRI/BBSRC AI for Drug Discovery Progamme, and successful candidates will join a cohort of students working on complementary projects in the AI for Drug Discovery space.

We are looking for highly motivated individuals who are passionate about contributing to new discoveries in drug discovery bioscience through the application of the latest techniques in AI and data science. Ideal candidates will have a grounding in both a natural science and data science, e.g. through a Master's degree or work experience in a subject such as bioinformatics or computational chemistry. Alternatively, you may have, for example, a first-class degree in computer science followed by biochemistry experience, or vice versa (qualifications and evidence thereof must be obtained before October 2025). You will be confident in performing data wrangling and analysis in a language such as Python, R or C++. Effective communication skills are essential.

We particularly encourage students from groups that are currently underrepresented in postgraduate science research, including black and minority ethnic students and those from a socio-economically disadvantaged background.

This studentship award can cover full tuition fees, UKRI stipend (currently £21,637) and a consumable allowance for a period of 4-years (pro-rata for part-time). 

Visit the Apply Pages for details on how to submit an application

Supervisors:

Professor Claudia Langenberg - Director, Precision Healthcare University Research Institute, QMUL

Project Partner: GSK

Expanding Predictive Toxicity with Multimodal Models of High-Dimensional Biology

Toxicity remains challenging for drug development with 30% of clinical trials failing due to toxicity. Models of toxicity are typically chemical structure and limited by the availability or biological relevance of data. The development of scalable biological assays across high-dimensional data layers (e.g. phenomics, transcriptomics, proteomics, metabolomics) has led to the availability of massive information-rich datasets. The aim of this project is to transform drug toxicity prediction by combining multimodal, high-dimensional biological data (e.g., phenomics, transcriptomics, proteomics, metabolomics) with chemical structure information using advanced AI methods to improve the accuracy, generalisability and availability of toxicity predictions.

We are seeking a highly motivated PhD student with a strong foundation in both molecular biology and data science, and a keen interest in toxicology. This could be demonstrated through a Masters degree in a relevant field such as bioinformatics, or alternatively, a first-class degree in computer science with subsequent bioscience experience, or vice versa. The ideal candidate will be proficient in Python coding, with experience in data wrangling, statistics, and machine learning. They will be passionate about applying AI to large biomolecular datasets, particularly within the context of toxicology and the use and analysis of high dimensional data latent spaces, to advance biological knowledge. This interdisciplinary project offers the opportunity to work at the cutting edge of AI and computational biology, bridging advanced data science techniques with meaningful biological and clinical insights in both industrial and academic research environments.

The project is in partnership with Recursion, a clinical stage techbio company that is industrializing large scale data generation across multiple biology data layers to decode biology.  Powerful AI models are used to embed this large-scale data generation and uniquely apply them to drug discovery problems. You will work closely with the world-class scientists at Recursion throughout the project and have access to our predictive chemistry, ADME and toxicology teams for advice and support.  Within QMUL, you will work at the Barnes lab and benefit from the team’s expertise in data science applied to AI and Drug Discovery.

Eligibility and Applying 

This project is part of the UKRI/BBSRC AI for Drug Discovery Progamme, and successful candidates will join a cohort of students working on complementary projects in the AI for Drug Discovery space.

We are looking for highly motivated individuals who are passionate about contributing to new discoveries in drug discovery bioscience through the application of the latest techniques in AI and data science. Ideal candidates will have a grounding in both a natural science and data science, e.g. through a Master's degree or work experience in a subject such as bioinformatics or computational chemistry. Alternatively, you may have, for example, a first-class degree in computer science followed by biochemistry experience, or vice versa (qualifications and evidence thereof must be obtained before October 2025). You will be confident in performing data wrangling and analysis in a language such as Python, R or C++. Effective communication skills are essential.

We particularly encourage students from groups that are currently underrepresented in postgraduate science research, including black and minority ethnic students and those from a socio-economically disadvantaged background.

This studentship award can cover full tuition fees, UKRI stipend (currently £21,637) and a consumable allowance for a period of 4-years (pro-rata for part-time). 

Visit the Apply Pages for details on how to submit an application. 

Supervisors:

Professor Michael Barnes - Professor of Bioinformatics, William Harvey Research Institute, QMUL

Dr Stephen MacKinnon - Vice President of Applied ML, Recursion

Dr Andrew Wedlake - Associate Director of Data Science, Recursion

Project Partner: Recursion

Interpretable Multi-Scale Integration Framework for Biological Systems: A Mechanistic Approach

Background & Significance 

AI-driven foundation models have the potential to transform biological research, excelling at different scale such as: 

• Molecular (protein structures, interactions)
• Genetic (gene expression, DNA regulation)
• Cellular (morphology, phenotypes, behaviour) 

Inspired by https://arxiv.org/pdf/2412.06993 

However, current models face two key challenges: 

1. Lack of cross-scale integration – Missing interactions across molecular, genetic, and cellular levels. 
2. Limited interpretability – Predictions remain black-box, reducing trust and usability. 

This project aims to develop an interpretable, multi-scale integration framework to enhance biological understanding, with applications in drug discovery and disease research. 

Research Objectives 

• Interpretable Multi-Scale Integration – Develop a framework to unify biological models, enhance interpretability, and resolve inconsistencies.
• Cross-Scale Information Flow – Enable bidirectional data exchange, optimize model contributions, and validate predictions.
• Dynamic Model Orchestration – Design autonomous agents for model integration, implement feedback loops, and ensure biological relevance.
• Interpretable Digital Cell Prototype – Build a multi-scale simulation, validate it, and use mechanistic insights to generate testable hypotheses. 

Impact 

This research will advance computational biology and AI interpretability by: 

• Enabling cross-scale integration of biological data.
• Improving mechanistic understanding through interpretable models.
• Enhancing trust and reliability in AI-driven biological research. 

This project is in conjunction with MSD.

Eligibility and Applying 

This project is part of the UKRI/BBSRC AI for Drug Discovery Progamme, and successful candidates will join a cohort of students working on complementary projects in the AI for Drug Discovery space.

We are looking for highly motivated individuals who are passionate about contributing to new discoveries in drug discovery bioscience through the application of the latest techniques in AI and data science. Ideal candidates will have a grounding in both a natural science and data science, e.g. through a Master's degree or work experience in a subject such as bioinformatics or computational chemistry. Alternatively, you may have, for example, a first-class degree in computer science followed by biochemistry experience, or vice versa (qualifications and evidence thereof must be obtained before October 2025). You will be confident in performing data wrangling and analysis in a language such as Python, R or C++. Effective communication skills are essential.

We particularly encourage students from groups that are currently underrepresented in postgraduate science research, including black and minority ethnic students and those from a socio-economically disadvantaged background.

This studentship award can cover full tuition fees, UKRI stipend (currently £21,637) and a consumable allowance for a period of 4-years (pro-rata for part-time). 

Visit the Apply Pages for further details on how to submit an application. 

Supervisors:  

Professor Venet Osmani - Professor of Clinical AI and Machine Learning, Digital Environment Research Institute, QMUL

Dr Kirill Shkura - Systems Biologist and Data Scientist, MSD

Soumya Ghosh - Director of Machine Learning, MSD

Project Partner: MSD

Development of Generalisable Multi-Modal ML Models for Digital Pathology

Bio-AI Health has a strong focus on development of ML based models that can help in patient screening and selection for clinical development.  Our models tend to focus on detection of key biomarkers of interest which are typically detected via some type of molecular or chemical test on a digital H&E image.  There are numerous publications which show the potential of such technology, however most of these models lack enough generalization for them to be successfully incorporated in real world practice.  Models tend to perform well when trained on specific data sets but fail when transferred to new labs or new cohorts.   

Our aim is to leverage molecular (or other) signals from these biomarkers and integrate them into our computer vision models in such a way that will allow for the model to focus on the signal of interest and lose any staining or processing bias.  This will allow us to train a generalizable model that can be effective across multiple labs / cohorts of data.  Additionally, our aim is to leverage the images plus molecular data for the development of the model, therefore vectorizing the molecular signal in such a way that we will be able to test and run the model on image data only.  Use of technology such as 10x Visium will allow for precise characterization of signal with a spatial component.  

The main objectives for the PhD project will be to:

• Model generalization across cohorts for digital pathology
• Multimodal integration of digital images and molecular signals generated from a RNAseq (or other omic data)
• Leverage spatial transcriptomics data and combine with computer vision for model development and optimization
• Help come up with innovative approaches for development of robust generalizable models on digital pathology data for personalized medicine
• Help with the integration of foundational pathology models for the fine tuning of specific tasks that allows for model generalization 

Candidates will need a strong background in deep learning, particularly computer vision, and a knowledge of omics including transcriptomics. Familiarity with histopathology images, knowledge of model generalisation techniques and experience working with multimodal AI or graph neural networks will be advantageous.  

Eligibility and Applying 

This project is part of the UKRI/BBSRC AI for Drug Discovery Progamme, and successful candidates will join a cohort of students working on complementary projects in the AI for Drug Discovery space.

We are looking for highly motivated individuals who are passionate about contributing to new discoveries in drug discovery bioscience through the application of the latest techniques in AI and data science. Ideal candidates will have a grounding in both a natural science and data science, e.g. through a Master's degree or work experience in a subject such as bioinformatics or computational chemistry. Alternatively, you may have, for example, a first-class degree in computer science followed by biochemistry experience, or vice versa (qualifications and evidence thereof must be obtained before October 2025). You will be confident in performing data wrangling and analysis in a language such as Python, R or C++. Effective communication skills are essential.

We particularly encourage students from groups that are currently underrepresented in postgraduate science research, including black and minority ethnic students and those from a socio-economically disadvantaged background.

This studentship award can cover full tuition fees, UKRI stipend (currently £21,637) and a consumable allowance for a period of 4-years (pro-rata for part-time). 

Visit the Apply Pages for further details on how to submit an application. 

Supervisors: 

Dr Vivek Singh - Lecturer in Digital Pathology, Barts Cancer Institute, QMUL

Dr Alan Jerusalmi - Chief Innovation Officer & Co-Founder, Bio-AI Health

Project Partner: Bio-AI Health