An AI-powered knowledge-graph approach for predicting biological threat responses

About the project

Interested in applying AI to solve real-world biological and security challenges? This project will bring together systems biology, machine learning and network science to build predictive models of host-pathogen interactions. You will develop cutting-edge AI approaches that could transform how we respond to novel infectious and biological threats.

The human body is a complex adaptive and hierarchical system of nested biological networks. These networks continuously respond to a multitude of external and internal signals, ranging from pathogen (bacterial, viral) exposure and environmental stressors to therapeutic interventions.

The ability to predict how biological networks respond to external perturbations has important applications across healthcare, defence and biosecurity. Biological threats, whether naturally occurring, engineered, or previously unknown, represent a significant risk to military personnel, critical infrastructure, and civilian populations. Rapidly understanding how a novel biological agent interacts with the human host and how the host networks respond is therefore a major challenge.

This project focuses on developing tools for prediction of how these networks respond to biological threats such as bacterial and viral pathogens.

Knowledge graphs provide a powerful framework for integrating, representing and interrogating heterogeneous information. Biomedical knowledge graphs integrate diverse molecular and disease-related information and are comprised of 100s of thousands of entities and millions of relations between those entities.

The project will first construct an integrated multi-omics knowledge graph incorporating host-pathogen interactions, molecular pathways, diseases, phenotypes and therapeutic interventions.

We will then develop graph foundation models and explainable AI methods to quantify how biological threats perturb molecular networks, identify pathways associated with disease severity and resilience, and predict potential intervention targets.

These approaches will provide a mechanistic framework for modelling biological responses and supporting preparedness against emerging infectious and environmental threats.

Lead supervisor

• Lead supervisor: Professor Rob Ewing BSc(Hons), DPhil Professor of Bioinformatics

Professor Rob Ewing BSc(Hons), DPhil

• Professor Rob Ewing BSc(Hons), DPhil: Professor of Bioinformatics

Professor Yihua Wang

• Cell Signalling in Disease
• Epithelial-Mesenchymal Crosstalk
• Pulmonary Fibrosis

Dr Zehor Belkhatir

• Design of estimation and control techniques, with convergence guarantees, for finite and infinite-dimensional systems
• Robust predictive modelling and analysis of high-dimensional data (e.g., network, time-series, imaging)
• The cross-over between artificial intelligence and mathematics for interpretable and robust computational algorithmic designs

Entry requirements

You must have an undergraduate degree (at least UK 2:1 honours degree) or its international equivalent in one of the following:

We welcome applicants with an interest in applying computational approaches to biological questions.

An interest in systems biology, host-pathogen interactions, AI or biological knowledge graphs would be welcomed.

Familiarity with, or experience in, any combination of bioinformatics, biological data analysis, programming, machine learning, network analysis or genomics would be useful, but applicants are not expected to have experience in all of these areas.

Relevant specialist skills can be developed during the PhD training.

If you have experience in any combination of these areas and are interested in developing your skills across biological and computational research, we encourage you to apply.

• biological sciences
• bioinformatics
• computer science
• data science
• or a related discipline

Fees and funding

For candidates looking to complete a PhD or EngD full-time we offer an enhanced tax-free stipend of up to £30k per year, along with a substantial budget for research, travel, and centre activities.

CISDnS is committed to promoting equality, diversity and inclusivity. We welcome all applicants regardless of their gender, ethnicity, disability, sexual orientation or age, and will give full consideration to applicants seeking flexible working patterns and those who have taken a career break or are transitioning into a new role. The University has a generous maternity policy, onsite childcare facilities, and offers a range of benefits to help ensure employees’ well-being and work-life balance.

How to apply

Apply now

Please apply via the online portal and select:

Applications should include:

• Academic year: 2027/28
• If you will be full time or part time
• Faculty: Engineering and Physical Sciences
• Search for "complex integrated systems"
• Select either PhD or EngD – if you are unsure, select PhD
• Add the name of the supervisor in section 2 of the application.
• your CV (resumé)
• 2 academic references
• degree transcripts and certificates to date

Faculty of Engineering and Physical Sciences

For questions about applying, email our Doctoral College (doctoralcollege-admissions@soton.ac.uk) .

Project leader

For an initial conversation, email Professor Rob Ewing (rob.ewing@soton.ac.uk) .

Research areas