4-year PhD fellowship in the Research Programme - Deep Learning-Accelerated Crystallography Pipeline

We welcome applications from candidates with a broad range of academic backgrounds and experiences for a 4-year PhD project on Mathematical and Machine Learning Aspects in Crystallography at Durham University.

Project description

This fellowship forms part of the Novo Nordisk Foundation-funded project “Deep Learning-Accelerated Crystallography Pipeline”, a collaboration between Durham University, the University of Copenhagen and the MAX IV synchrotron. You will work with an international team of mathematicians, crystallographers and data scientists. The project aims to transform small-molecule structure determination by developing mathematical methods and integrating machine learning into crystallographic workflows. The successful candidate will develop theoretical and computational approaches to improve structure solution, refinement and validation.

Supervisors

• Principal Supervisor: Professor Norbert Peyerimhoff, Department of Mathematical Sciences, Durham
• Co-Supervisor: Dr Niklas Ruth, Advanced Research Computing (ARC), Durham

Start and duration

The PhD commences on 1 January 2027 or as soon as possible thereafter and runs for four years.

Work environment

You will be based in the Department of Mathematical Sciences at Durham University. The project includes close collaboration with Professor Anders Østergaard Madsen, Principal Investigator, at the University of Copenhagen and Dr Lennard Krause at MAX IV in Lund. Your supervisors provide complementary expertise in mathematics, modern crystallography and machine learning, with opportunities to interact with OlexSys.

Job description

Your key tasks are:

• Carrying out an independent research project under supervision, including deriving its mathematical basis and implementing results in code;
• Completing PhD courses or equivalent training;
• Participating in research seminars;
• Disseminating results through visits, workshops and conferences;
• Writing a PhD thesis.

Key criteria for applicants

Applicants should have:

• A qualification equivalent to a Master’s degree in Chemistry, Mathematics or Computer Science by the start of the PhD;
• A curious mind-set and strong background in quantum crystallography and its underlying mathematical aspects;
• Adequate programming skills and willingness to apply machine learning where required;
• Optionally, previous experience contributing to open-source scientific software.

Terms of employment and stipend

The position is fixed-term for four years. The stipend covers the standard UK salary and all university/institutional fees, with a budget for travel, conferences and equipment.

How to apply and application deadline

Please submit a single PDF containing:

• A CV (max. 2 pages);
• A cover letter (max. 1 page) describing your motivation;
• A certified copy of your Master’s diploma and transcript, with an authorised English translation where required. If the degree is not yet completed, a certified/signed recent transcript or written statement from the institution or supervisor is accepted;
• The names, institutions, positions and e-mail addresses of up to two potential referees, who may be contacted for shortlisted candidates;
• Optionally, a link and username for a repository with your contributions to publicly available scientific software.

Applications should be sent to norbert.peyerimhoff@durham.ac.uk and paul.n.ruth@durham.ac.uk.

The application deadline is 7th of October, 2026, 23:59 BST (British Summer Time).