AstraZeneca taps AI for drug discovery in deal with Berg

Reuters  |  LONDON 

(Reuters) - has forged a research collaboration with Boston-based Berg, a specialist in artificial intelligence for drug hunting, in the latest sign of big pharma's interest in using supercomputers for drug discovery.

The tie-up will focus on finding and evaluating novel ways of treating Parkinson's disease and other neurological disorders.

Financial terms were not disclosed but the deal involves providing with chemical fragments that the U.S. company will search for potential drug candidates.

Berg's scientists have already used to find novel biological targets for new medicines by comparing detailed data from tissue samples collected from diseased and healthy individuals.

will have the right to secure an exclusive licence to any of the drug candidates coming out of the work.

A growing number of big pharmaceutical companies, including GlaxoSmithKline, Sanofi and Merck are exploring the potential of through alliances with start-ups.

The aim is to harness modern computing systems to predict how molecules will behave and how likely they are to make a useful drug, thereby saving time and money on unnecessary tests.

"2017 has really been the year when has been taken seriously and we are now seeing early signs of adoption and implementation in the broader industry," Niven Narain, chief executive of privately owned Berg, told "In prior years there was some scepticism."

For Berg, which also has its own experimental cancer drugs developed using in clinical trials, the deal with is its first major pharma collaboration.

Narain expects to announce more partnerships soon. "We have a few that are very close to being finalised," he said.

Other young companies in the drug discovery sector include Britain's BenevolentAI and Exscientia, and U.S. based firms Numerate, twoXAR, Atomwise and InSilico Medicine.

(Reporting by Ben Hirschler; Editing by Elaine Hardcastle)

(This story has not been edited by Business Standard staff and is auto-generated from a syndicated feed.)