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
- Evotec and Rappta Therapeutics Enter Discovery and Development Partnership Focused on Oncology Targeton November 24, 2020 at 5:01 pm
November 24, 2020 (ACCESSWIRE via COMTEX) -- EVOTEC TO SUPPORT RAPPTA IN DEVELOPING SMALL MOLECULES THAT REACTIVATE PP2A, A KEY TUMOR SUPPRESSORTHE COLLABORATION LEVERAGES EVOTEC'S ONCOLOGY EXPERTISE ...
- Global Cancer Immunotherapy Drug Discovery Outsourcing Market 2020 Size, Market Share, Key Players, Segmentation Development and Forecast by 2026on November 24, 2020 at 2:03 am
MarketsandResearch.biz has newly published research study entitled Global Cancer Immunotherapy Drug Discovery Outsourcing Market 2020 by Company, Regions, Type and Application, Forecast to 2026 that ...
- Schrödinger inks multi-target drug discovery deal with Bristol-Myers Squibbon November 23, 2020 at 8:52 am
(NASDAQ:SDGR) jumps 7% premarket on collaboration with Bristol-Myers Squibb (NYSE:BMY) to discover, develop, and commercialize therapeutics in multiple disease areas.
- The peptides and macrocycle drug discovery services market is estimated to be worth USD 1.5 billion in 2030, predicts Roots Analysison November 23, 2020 at 8:49 am
Roots Analysis has announced the addition of “ Peptides and Macrocycle Drug Discovery: Services and Platforms Market, 2020-2030 ” report to its list of offerings. The report features an extensive ...
- Schrödinger Announces a Multi-Target Drug Discovery, Development, and Commercialization Collaboration with Bristol Myers Squibbon November 23, 2020 at 3:30 am
Schrödinger, Inc., whose physics-based software platform is transforming the way therapeutics and materials are discovered, today announced a discovery collaboration with Bristol Myers Squibb Company ...
- Understanding ion channel inhibition to open doors in drug discoveryon November 23, 2020 at 2:03 am
Scientists have discovered how drug-like small molecules can regulate the activity of therapeutically relevant ion channels—and their findings could transform ongoing drug development efforts.
- In silico drug discovery services market is likely to be worth USD 124 million by 2030, growing at an annualized rate of 15.6%, claims Roots Analysison November 21, 2020 at 2:12 pm
Roots Analysis has announced the addition of “In Silico / Computer-Aided Drug Discovery Services Market: Focus on Large Molecules (Antibodies, Proteins, Peptides, Nucleic Acid, Gene Therapy and ...
- AI drug discovery platform AbCellera files for a $200 million IPOon November 20, 2020 at 2:00 pm
A bCellera Biologics, which provides an AI powered drug discovery platform for antibody therapies, filed on Friday with the SEC to raise up to $200 million in an initial public offering. The company ...
- Insights on the AI in Drug Discovery Global Market to 2026 - Opportunity Analysis and Industry Forecaston November 20, 2020 at 9:00 am
The "AI in Drug Discovery Market by Component, Technology, by Application, by End User, by Geography - Global Opportunity Analysis ...
- Moving targets in drug discoveryon November 19, 2020 at 7:49 am
Drug Discovery is a lengthy and costly process and has faced a period of declining productivity within the last two decades resulting in increasing importance of integrative data-driven approaches. In ...
via Bing News