Digital technology collaboration to develop machine learning for sustainable drug discovery – Mirage News


Scientists are using the latest Digital technology to support the development of machine learning models of sustainable chemistry that can be used in the early phase of drug discovery.

The University of Nottingham’s School of Chemistry is extending their collaboration with DeepMatter (AIM: DMTR), the AIM-quoted company focusing on digitising chemistry to provide access to their DigitalGlassware® platform to support the Universities’ sustainable chemistry initiatives. This collaboration will focus on the development of machine learning models of sustainable chemistry for researchers in the pharmaceutical sector and related chemical-based industries.

The project will be led by Professor Jonathan Hirst, Professor of Computational Chemistry & Royal Academy of Engineering Chair in Emerging Technologies, in the University’s School of Chemistry, and will see scientists use DigitalGlassware® to develop machine learning techniques to help chemical engineers and chemists make their manufacturing processes more sustainable. In an extension to the project, Professor Hirst will work with scientists at the University of Nottingham’s Centre for Sustainable Chemistry, to build interactive machine learning models of sustainability, effectively rules to follow, that can be used in the early discovery phase by researchers in the pharmaceutical sector when seeking to develop new drugs and related chemicals in a more sustainable manner.

A further project, run by Professor Ross Denton, will look at capturing data in the lab to help with forward prediction using computational modelling.

DigitalGlassware® is an integrated software, hardware and artificial intelligence enabled platform, which allows chemistry experiments to be accurately and systematically recorded, coded and entered into a shared data cloud allowing real-time and post-hoc analysis of the chemistry. The systematic structuring and recording of the data within DigitalGlassware® means that the platform can provide large and structured chemical reaction data sets, suitable for interrogation by machine learning techniques in a way not previously possible with smaller sets of manually collected and recorded data.

At the moment, there are all kinds of inefficiencies, which are largely neglected in the search for new chemicals with specific desired properties. This project will provide machine learning tools for chemists and chemical engineers that will help them address two questions. How do we find greener synthetic routes to chemicals? And how do we identify greener target molecules from the outset? DigitalGlassware® is a great fit with this project as its machine learning and AI enabled approach to data delivers greater value and will help to revolutionise chemistry in both productivity gains and the discovery of new insights.

Mark Warne, CEO of DeepMatter Group, commented: “Having worked with the University of Nottingham previously in their Digital Teaching Laboratory, we are delighted to be working with Jonathan in his challenge to design, make and ultimately manufacture new molecules in a more sustainable fashion. We recognise that significant changes are required in the way science is organised and conducted for there to be progress towards a more sustainable environment and we are pleased that our technology will play a role in developing the industry’s sustainability. Jonathan, Ross and their teams will have access to our innovative cloud-based platform, DigitalGlassware®, allowing them to share and use the data with machine learning and AI technologies to provide unique perspectives.”

/Public Release. This material comes from the originating organization and may be of a point-in-time nature, edited for clarity, style and length. View in full here.