We are creating a unified UKRI website that brings together the existing research council, Innovate UK and Research England websites.
If you would like to be involved in its development let us know.

Simulation and machine learning techniques at STFC's Hartree Centre could improve the way we treat diseases

9 June 2020

Research carried out at STFC’s Hartree Centre could help reduce the cost and speed up the development of new types of drugs, such as for cancer therapies or new types of antibiotics.  The research, which focuses on a specific type of peptide, could improve the way we treat diseases in the future.

Peptides are short chains of amino acids that occur naturally in our body, plants or bacteria to control diverse functions. Examples of current peptide-based medicines in use today include the treatment of diabetes and obesity.  There are currently many other peptide-based medicines in development, however the time and cost of developing new drugs and taking them to market is soaring.

Recently, through their ongoing collaboration with IBM Research, scientists based at the STFC Hartree Centre have been working with leading researchers from AstraZeneca to gain deeper insights into a specific type of peptide, known as ‘cyclic peptides’. Cyclic peptides have their amino acid chains looped together head to tail.  Most importantly, they have the potential to bind with certain disease targets more successfully than small molecules. However, to do this a cyclic peptide needs to be permeable enough to cross the cell membrane and get inside of cells, which is a challenge.

The IROR Programme: Simulation and machine learning for future medicine

Credit: STFC/IBM Research

Using the powerful supercomputers at the Hartree Centre, at STFC’s Daresbury Laboratory at Sci-Tech Daresbury, the team developed computational workflows that blend molecular dynamics - a technique that simulates the movements of atoms and molecules - with machine learning. This enabled them to quickly and accurately study the shapes and patterns of movement of individual cyclic peptides, gaining new insights on how their shape influences permeability in the development of new drugs.

Anders Hogner, team leader of computational chemistry, at AstraZeneca, said: “Our collaboration with the Hartree Centre and IBM has provided us with valuable new insights into this challenging system, and a demonstration of exciting new tools for drug discovery. We are very excited to build upon this collaboration as we work towards large scale benchmarking exercises and the adoption of machine learning capabilities in new drug development.”

As part of the project, the team also created user-friendly ‘computational notebooks’. These allow scientists to access the power of supercomputers, without the need for specialist high performance computing expertise themselves, enabling them to carry out simulation, visualisation and analysis through a laptop.

This research was carried out as part of the Innovation Return on Research (IROR) programme, a collaboration between STFC’s Hartree Centre and IBM Research, with a mission to help UK industry solve industrial problems and reduce risk, creating economic and societal impact for the UK.

Alison Kennedy, Director of STFC’s Hartree Centre, said: “Using machine learning techniques is a fantastic way for industry to save time and costs associated with this type of research, which could improve the way we treat diseases in the future. By combining the technology and skills here at STFC’s Hartree Centre with those of IBM Research through the IROR programme, we are in a great position to help UK industry, and our academic community, find faster and cheaper solutions to their challenges. Most importantly, the outcomes of this particular work could be transferable for use by other researchers in the field, saving costs and boosting productivity, the benefit of the UK economy.”

Read further information about this research at the case study.

If you enjoyed this, never miss a story by subscribing to our newsletter FASCINATION:

Subscribe to Fascination

Last updated: 10 June 2020


Science and Technology Facilities Council
Switchboard: +44 (0)1793 442000