Researchers have designed a machine learning algorithm that predicts the outcome of chemical reactions with much higher accuracy than trained chemists and suggests ways to make complex molecules, removing a significant hurdle in drug discovery.
Our platform is like a GPS for chemistry
University of Cambridge researchers have shown that an algorithm can predict the outcomes of complex chemical reactions with over 90% accuracy, outperforming trained chemists. The algorithm also shows chemists how to make target compounds, providing the chemical ‘map’ to the desired destination. The results are reported in two studies in the journals ACS Central Science and Chemical Communications.
A central challenge in drug discovery and materials science is finding ways to make complicated organic molecules by chemically joining together simpler building blocks. The problem is that those building blocks often react in unexpected ways.
“Making molecules is often described as an art realised with trial-and-error experimentation because our understanding of chemical reactivity is far from complete,” said Dr Alpha Lee from Cambridge’s Cavendish Laboratory, who led the studies. “Machine learning algorithms can have a better understanding of chemistry because they distil patterns of reactivity from millions of published chemical reactions, something that a chemist cannot do.”
The algorithm developed by Lee and his group uses tools in pattern recognition to recognise how chemical groups in molecules react, by training the model on millions of reactions published in patents.
The researchers looked at chemical reaction prediction as a machine translation problem. The reacting molecules are considered as one ‘language,’ while the product is considered as a different language. The model then uses the patterns in the text to learn how to ‘translate’ between the two languages.
Using this approach, the model achieves 90% accuracy in predicting the correct product of unseen chemical reactions, whereas the accuracy of trained human chemists is around 80%. The researchers say that the model is accurate enough to detect errors in the data and correctly predict a plethora of difficult reactions.
The model also knows what it doesn’t know. It produces an uncertainty score, which eliminates incorrect predictions with 89% accuracy. As experiments are time-consuming, accurate prediction is crucial to avoid pursuing expensive experimental pathways that eventually end in failure.
In the second study, Lee and his group, collaborating with the biopharmaceutical company Pfizer, demonstrated the practical potential of the method in drug discovery.
The researchers showed that when trained on published chemistry research, the model can make accurate predictions of reactions based on lab notebooks, showing that the model has learned the rules of chemistry and can apply it to drug discovery settings.
The team also showed that the model can predict sequences of reactions that would lead to a desired product. They applied this methodology to diverse drug-like molecules, showing that the steps that it predicts are chemically reasonable. This technology can significantly reduce the time of preclinical drug discovery because it provides medicinal chemists with a blueprint of where to begin.
“Our platform is like a GPS for chemistry,” said Lee, who is also a Research Fellow at St Catharine’s College. “It informs chemists whether a reaction is a go or a no-go, and how to navigate reaction routes to make a new molecule.”
The Cambridge researchers are currently using this reaction prediction technology to develop a complete platform that bridges the design-make-test cycle in drug discovery and materials discovery: predicting promising bioactive molecules, ways to make those complex organic molecules, and selecting the experiments that are the most informative. The researchers are now working on extracting chemical insights from the model, attempting to understand what it has learned that humans have not.
“We can potentially make a lot of progress in chemistry if we learn what kinds of patterns the model is looking at to make a prediction,” said Peter Bolgar, a PhD student in synthetic organic chemistry involved in both studies. “The model and human chemists together would become extremely powerful in designing experiments, more than each would be without the other.”
The Latest on: Drug discovery
via Google News
The Latest on: Drug discovery
- Russian Doping Chief Says Thousands of Drug Tests Were Manipulatedon October 14, 2019 at 3:12 pm
COLORADO SPRINGS — Russia made thousands of changes to the drug-test results of an unspecified number of ... sports — a return to the pariah status that followed the 2015 discovery — if its ...
- Week In Review: Jiangsu's CTFH Signs $200-Million AI Drug Discovery Pact With Insilicoon October 13, 2019 at 11:31 pm
Jiangsu Chia Tai Fenghai Pharma (CTFH) signed a $200-million agreement with Insilico Medicine, an AI drug discovery company headquartered in Hong Kong, to discover two triple-negative breast cancer ...
- Drug Discovery Informatics Market: Global Key Players, Trends, Share, Industry Size, Growth, Opportunities, Forecast To 2026on October 10, 2019 at 3:53 am
As per Wise Guy report, due to the projected price of research and development for medical discovery, the drug discovery computer industry is projected to record 15.3 percent CAGR over the forecast ...
- Drug Discovery Informatics Market 2019 Global Trend, Segmentation And Opportunities Forecast To 2024on October 9, 2019 at 12:29 pm
Company profile includes assign such as company summary, financial summary, business strategy and planning, SWOT analysis and current developments. The Global Drug Discovery Informatics Market is ...
- Drug Discovery Informatics Market 2019: Global Key Players, Trends, Share, Industry Size, Segmentation, Opportunities, Forecast To 2026on October 9, 2019 at 1:40 am
A new market study, titled "Discover Global Drug Discovery Informatics Market Upcoming Trends, Growth Drivers and Challenges" has been featured on WiseGuyReports. As per Wise Guy report, due to the ...
- Scientists invent new technology to streamline drug discoveryon October 7, 2019 at 12:49 pm
The final step is when scientists use a mass spectrometer to identify the unpainted regions, which is where the proteins touch. Current technologies in early phase drug discovery, such as ...
- Evotec Reaches Milestone In Integrated Drug Discovery And Development Partnership With Aeovianon October 4, 2019 at 12:22 am
Aeovian and Evotec entered into a drug discovery partnership in 2017 to discover and develop new and highly selective mTORC1 inhibitors for the treatment of rare and age-related diseases. After ...
- Drug Discovery Technologies Market 2019|Top Key Players Analysis, Trends, Global Size Forecast To 2024on October 2, 2019 at 6:57 am
The scope also covers competitive overview providing company market shares along with company profiles for major revenue contributing companies. The global Drug Discovery Technologies market is ...
- New translational research centers to speed, diversify Alzheimer’s drug discoveryon October 2, 2019 at 6:00 am
The Alzheimer Centers for the Discovery of New Medicines are designed to diversify and reinvigorate the Alzheimer’s disease drug development pipeline. NIA is part of the National Institutes of Health.
- Xylyx Bio announces development of disease-specific cell culture products to accelerate anti-fibrotic drug discovery and screeningon October 2, 2019 at 5:25 am
Fibrosis-specific substrates provide a disease-relevant in-vitro testing platform to inform decision-making in early-stage drug discovery NEW YORK, Oct. 2, 2019 /PRNewswire/ -- Xylyx Bio, a pioneer in ...
via Bing News