Research Fellow in AI/Computational Structural Biology

About us

Biosciences is one of the world’s foremost centres for research and teaching in the biological sciences and one of the largest Divisions within UCL, undertaking a significant amount of research and teaching. The Division has a diverse portfolio addressing all areas of biology from protein interactions to cell function, organism development, genetics, population studies and the environment. Computational modelling approaches are frequently used alongside experimental research programmes and much of our research crosses traditional boundaries, including the relationship of biodiversity to the health of the planet. Activity is underpinned by high calibre science technology platforms and state of the art equipment. Educational activity includes a range of undergraduate programmes, an expanding number of Masters Programmes and a substantial number of postgraduate research students.

This is an exciting opportunity to join the Computational Biology Group headed by Professor Christine Orengo. The successful applicant will join a pioneering research effort focussed on developing novel AI-based/ML methods for identifying structure function groups (FunFams) in domain superfamilies and characterising their ligand binding pockets.

About the role

We are seeking a highly motivated researcher to develop artificial intelligence based novel algorithms and computational workflows to identify domain functional families (FunFams) in the CATH-TED database comprising relatives with highly similar structures and functions. This will also involve building workflows to map all available ligand data from ChEMBL and other public sources to these FunFams and to capture characteristics of the binding pockets in the family. The project is a collaboration with the groups of David Jones, also at UCL and Sameer Velankar who manages the PDBe-KB at the EBI. Workflows will also be developed to annotate protein structures in the PDBE-KB and AFDB with domain family information and information on binding pocket characteristics.

The researcher would be expected to have knowledge of protein structure, protein ligand binding, machine learning and expertise in workflow development. Information generated by the project will be widely disseminated via the TED, PDBeKB and AFDB websites.

This role is an open-ended contract with a funding end date of 14/08/2026 in the first instance.

A job description and person specification can be accessed at the bottom of this page.

If you need reasonable adjustments or a more accessible format to apply for this job online or have any queries about the application process, please contact Biosciences staffing at .

About you

You must have a PhD in bioinformatics, biosciences, computational biology, computer science, data science or a related subject area and proven knowledge of python programming, developing machine learning/AI based tools and HPC.

You will be expected to work as part of a tightly integrated team of computational biologists to produce novel algorithms and workflows to classify domain functional families and characterise their binding pockets.

In addition to developing and conducting the research, you will contribute expertise to the overall research effort in the Orengo group. You will also communicate results as scientific papers in leading journals, and as scientific presentations at national and international conferences.

Customer advert reference: B02-08858

Advert information