2026 Projects
Novel biomarkers for human gut AMPK activation using AI-driven multi-omics analysis
This PhD project will use AI-driven analysis of multi-omics data to discover blood-based biomarkers that report activation of AMPK, a key sensor of cellular “fuel deficiency” linked to obesity, diabetes and healthy ageing. The student will integrate metabolomic, proteomic, phosphoproteomic and transcriptomic datasets from human, animal and cell studies with known AMPK activators (e.g. metformin, salicylate, dietary compounds) to identify a robust molecular signature of AMPK activation. These features will be combined into an interpretable “AMPK Activation Score” and applied to samples from human dietary intervention studies to assess how foods and nutrients modulate energy metabolism in vivo. The project offers interdisciplinary training at the interface of AI, nutrition, metabolism and biomarker discovery.
Click here to access the application form
Machine learning-assisted antibacterial drug discovery
This PhD project will use machine learning to accelerate the discovery of new antibacterial drugs, focusing on plant-derived phytochemicals as a rich source of novel compounds. The student will develop predictive models that link chemical structure to antibacterial activity against multidrug-resistant (MDR) bacteria, helping to prioritise promising candidates and potentiators of existing antibiotics. These predictions will be tested experimentally using microbiology-based phenotypic assays and analytical chemistry for compound characterisation. Working at the interface of AI, microbiology and natural-product chemistry, the project aims to deliver both new methods and new leads for tackling antimicrobial resistance.
Click here to access the application form
AI-Driven Multi-Omics Biomarker Discovery and Early Detection of Mastitis in Dairy Cattle
This PhD project will use explainable AI and multi-omics integration to enable early detection of subclinical mastitis in dairy cattle. The student will combine mid-infrared milk spectra, transcriptomics, genotype data and routine milk records to identify robust molecular biomarkers and build interpretable machine-learning models for mastitis risk prediction. The work aims to deliver biologically meaningful biomarker panels and decision tools that improve herd health, reduce antimicrobial use and support precision livestock management. The project offers interdisciplinary training spanning bioinformatics, AI, microbiology and animal health, including an industrial placement at AFBI.
Click here to access the application form
Using AI-driven computer vision and high-throughput multiplexed human cell infection assays to dissect intra- and intercellular signalling networks and discover new antimicrobials
This PhD project will combine AI-driven computer vision with high-throughput, multiplexed human cell infection assays to understand how host cells and bacteria interact at single-cell resolution. The student will engineer visually barcoded libraries of mammalian cells and bacteria, track their interactions by high-content live fluorescence imaging, and develop machine-learning pipelines to decode complex intra- and intercellular signalling networks during infection. By linking specific host and bacterial phenotypes to infection outcomes and drug responses, the project will identify candidate targets for new antibiotics and host-directed therapies, followed by proof-of-concept testing with existing inhibitors.
Click here to access the application form.
AI-enabled antigen discovery for vaccines targeting One-Health antimicrobial-resistant pathogens
This PhD project will harness AI to discover new vaccine antigens against antimicrobial-resistant (AMR) pathogens in a One-Health framework. Working with uniquely rich experimental datasets that link real immune responses and protection outcomes across multiple AMR bacteria, the student will develop interpretable machine-learning models to identify “protection signatures” and high-value vaccine targets. These predictions will then be tested in the lab using Streptococcus infection models, providing an iterative loop between AI and experiment. The project offers interdisciplinary training across computational biology, infection immunology and translational vaccine research, with industry exposure via the QUB spin-out VacTimmune.
Click here to access the application form
Harnessing AI to Define Extracellular Vesicle-Mediated Immune Regulation
This PhD project will use advanced AI and machine learning to uncover how extracellular vesicles (EVs) from mesenchymal stem cells regulate immune cell behaviour. The student will generate multi-omics datasets by profiling the molecular cargo of EVs and measuring how dendritic cells, B cells and T cells respond to EV treatment at the single-cell level. They will then develop and apply models (e.g. graph neural networks and multimodal autoencoders) to link EV composition to specific immune outcomes and identify key regulatory molecules or cargo combinations. Ultimately, the project aims to move from descriptive correlations to mechanistic understanding of EV-mediated immune regulation, informing future cell-free immunotherapies.
Click here to access the application form.
HAPPI: Harnessing AI for personalised nutrition to promote healthy ageing in older adults
This PhD project, HAPPI, will harness AI to develop personalised nutrition solutions that help older adults prevent or reverse undernutrition and maintain independence. The student will use machine learning on large ageing cohorts (including clinical trial and NICOLA data) to identify diet–function patterns and build transparent algorithms that generate tailored diet plans. They will then co-design and test a “Healthy Ageing Diet” mobile app with older adults, creating an AI-driven tool that supports long-term dietary behaviour change and improved health and functional outcomes.
Click here to access the application form.
Contextual Biological Modelling for Phenotypic Prediction
This PhD project will develop contextual probabilistic “world models” to understand and predict how cancer systems behave differently across contexts such as cell type, treatment, disease state and genetic background. Using the Contextual Probability (CP) framework, the student will model multiple biological contexts jointly, identify mechanisms that are stable or rewired between them, and build context-aware predictors of phenotypic responses. The work will focus on p53-family signalling in colorectal cancer (CRC) as a case study, using multi-omics and perturbation datasets to address a specific CRC prediction problem and generate new, testable hypotheses about context-dependent behaviour. The project offers interdisciplinary training in AI for bioscience, mathematical modelling, systems biology and computational cancer research.
Click here to access the application form.