If you’re interested in AI for biotechnology, multi-omics, and real-world industrial impact, this fully funded PhD studentship at Teesside University (Centre for Digital Innovation) offers a genuinely exciting research problem: using transformer-based deep learning to integrate imaging + omics + modelling to anticipate harmful fermentation changes before they become costly.
Author: Dr Niaz Chowdhury (LinkedIn)
Designation: Lecturer (Computer Science)
Affiliation: Ulster University, Birmingham, UK
Scholarship snapshot
- Project title: Transformer-Bio: integrating imaging and omics in large-scale food fermentation (visit this link to apply: Apply now)
- Studentship code: PHD
- Supervisor: Professor Claudio Angione
- Research centre: Centre for Digital Innovation
- Deadline: 07/01/2026 (5:00 PM)
- Start date: October 2026 (expected)
Why this project matters
Industrial food fermentation isn’t just about taste — it’s also about consistency, cost, waste, and emissions.
In continuous flow systems of Fusarium venenatum (used by Quorn, one of the leading meat-substitute companies), the fermentation can produce mutant strains known as C-variants. These variants develop altered hyphal branching patterns, which are:
- Very difficult to detect early
- Capable of changing product texture (often making it crumbly)
- A common reason fermentations get terminated early, causing high costs and avoidable emissions
One major challenge remains: it’s not yet clear why these branching patterns emerge, which makes proactive prevention difficult.
What the PhD will do
This project aims to build a deep learning method that can integrate multiple data sources to predict the onset of problematic branching patterns.
You will work on:
- Multi-omic characterisation of the fermentation process (to help explain why changes occur)
- Integration of metabolic modelling, imaging, and omics data
- A transformer-based architecture designed for time-resolved, multi-modal data (a strong fit for complex biological systems)
Recent research has shown transformers can perform strongly in tasks such as multi-omics and imaging data imputation and integration, which makes this approach a promising direction for fermentation prediction and monitoring.
What makes it especially strong
Industry-linked, real-world problem
This isn’t a theoretical “toy” dataset. The project addresses a concrete industrial challenge in large-scale food fermentation, with clear value for quality assurance and sustainability.
Interdisciplinary team and newly collected data
You’ll be part of a multidisciplinary group spanning:
- Computer science and deep learning
- Omics and molecular biology
- Fermentation and modelling
The data will be newly collected by the wider project team (including lab technician support), giving you the chance to shape the analysis pipeline from the ground up.
Wider collaboration network
Alongside Prof Angione’s team, you’ll work with:
- Prof Peter O’Toole (University of York)
- Prof Safwan Akram (National Horizons Centre)
- Prof Annalisa Occhipinti (Teesside University)
- Dr Nanda Puspita (Quorn)
Funding and eligibility
What the funding covers
This is a fully funded PhD studentship, covering:
- Tuition fees for up to four years (full-time registration)
- Annual tax-free stipend: £20,780 for four years (subject to satisfactory progress)
- Requirement: you must complete within four years
Who can apply
Applications are welcome from UK and international students.
Academic requirements:
- A 2:1 or above (or expected) in computer science, biological, chemical, physical sciences, or mathematics
- A Master’s in a relevant discipline is desirable but not essential
- A demonstrable understanding of the research area is expected
International applicants:
- A limited number of bursaries may be available to support full studentships for outstanding international candidates (highly competitive)
- Usual requirements apply: English language criteria, ATAS clearance, and visa processes where relevant
The wider doctoral training programme emphasises inclusive recruitment and welcomes applicants from underrepresented backgrounds, valuing curiosity, creativity, and commitment alongside conventional achievement.
How to apply
To apply, you’ll need to submit an Expression of Interest via the designated form for the project(s) you choose.
- You can apply for up to two projects within the scheme (even across different universities)
Academic enquiries:
For questions about the research, contact C.Angione@tees.ac.uk.
Key dates (for your diary)
- Application deadline: 07/01/2026 (5:00 PM)
- Interviews: expected February 2026 (date to be confirmed)
- Start: October 2026
What a strong application might highlight
If you’re serious about standing out, consider showing evidence of:
- Comfort with Python and modern deep learning workflows (e.g., PyTorch / TensorFlow)
- Understanding of transformers and sequence/time-resolved modelling
- Interest (and ideally experience) in omics, imaging, bioinformatics, or computational biology
- Ability to work across disciplines and communicate with non-computer-scientists
- Motivation connected to industrial impact, sustainability, and food systems
Even if your background is more biology-heavy or more CS-heavy, this project can suit you if you can demonstrate the willingness and ability to bridge the gap.
Final thought
Transformer-Bio sits right at the intersection of AI innovation and biotech industry impact. If you want a PhD where your work could help prevent real production failures and reduce waste, while developing cutting-edge multi-modal transformer methods, this is one to take seriously.
