Researchers develop new AI-driven model for drug discovery

Researchers at the University of Cambridge have developed a new artificial intelligence (AI) algorithm, which can predict the complex chemical reactions outcomes with more than 90% accuracy and facilitate the drug discovery by suggesting ways to make complex molecules. Providing the chemical ‘map’ to the desired destination, the new algorithm also shows chemists how to […]

The post Researchers develop new AI-driven model for drug discovery appeared first on PharmaNewsDaily.com.

Researchers at the University of Cambridge have developed a new artificial intelligence (AI) algorithm, which can predict the complex chemical reactions outcomes with more than 90% accuracy and facilitate the drug discovery by suggesting ways to make complex molecules.

Providing the chemical ‘map’ to the desired destination, the new algorithm also shows chemists how to make target compounds, a challenge in drug discovery and materials science.

The algorithm uses tools in pattern recognition to recognize how chemical groups in molecules react and then uses the patterns in the text to learn how to ‘translate’ between the two languages.

The model is said to be accurate enough to detect errors in the data and correctly predict a range of complex reactions.

University of Cambridge researchers develop develop new AI-driven model for drug discovery

University of Cambridge researchers develop develop new AI-driven model for drug discovery. Image courtesy of Gerd Altmann from Pixabay.

Dr Alpha Lee – lead researcher from Cambridge’s Cavendish Laboratory said: “Making molecules is often described as an art realised with trial-and-error experimentation because our understanding of chemical reactivity is far from complete.

“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 researchers have shown the practical potential of the method in drug discovery by collaborating with the biopharmaceutical company Pfizer during the second study.

The model can reduce the time of preclinical drug discovery by predicting sequences of reactions that would lead to the desired product.

“Our platform is like a GPS for chemistry. It informs chemists whether a reaction is a go or a no-go, and how to navigate reaction routes to make a new molecule,” added Dr Alpha Lee.

The results are reported in two studies in the journals ACS Central Science and Chemical Communications.

The research was supported by the Winton Programme for the Physics of Sustainability and the Herchel Smith Fund.

Related posts

The post Researchers develop new AI-driven model for drug discovery appeared first on PharmaNewsDaily.com.

Related Posts

CATEGORIES
Share This