Using AI and Machine Learning to Speed-Up Drug Development
Machine Learning set to create quality improvements in manufacturing processes
A neural network algorithm developed by STFC Scientific Computing’s AI for Science theme is being used by CrystalGrower Ltd, a spin out from the University of Manchester.
CrystalGrower clients in fields from pharmaceuticals to agrochemicals can now use the machine learning tool to analyse thousands of digitally simulated crystals faster and more accurately. This is enabling scientists to reach research insights in fields including drug development, chemicals and foods, far quicker than was previously possible.
CrystalGrower Ltd. is able to generate an infinite number and shapes of synthetic 3D crystals across many chemical systems for research purposes. However, analysing the results requires more scientists with specialist domain expertise than are available, so CrystalGrower approached STFC Scientific Computing to devise a novel way to auto cluster the crystals.
CrystalGrower chose to partner with STFC Scientific Computing (SC) to be able to access their extensive Artificial Intelligence (AI) and Machine Learning (ML) skills. SC provides ML and AI analysis services to all of STFC’s experimental facilities, as well as to external organisations.
“We chose to work with SC’s AI for Science Theme for its deep experience in the fundamental science side, as well as its impressive expertise in the fields of artificial intelligence and machine learning as applied to science. This has been a successful and productive collaboration.”
Dr. Adam Hill, Chief Scientific Officer, CrystalGrower Ltd
