Google DeepMind: Drug developers seek a structural advantage from AI

Receive free Lex updates

we will send you myFT Daily Digest Email summary of latest news Lex There is news every morning.

Dr. Google gets a bad rap. Researching symptoms online can be worrying for someone with cyberchondriasis. But it’s hard to criticize the Alphabet-owned company’s latest contribution to disease diagnosis. Google DeepMind’s new artificial intelligence tool can predict whether human genetic mutations are likely to be harmful. This should speed up the detection of diseases caused by rare genetic variants.

This achievement demonstrates the power of AlphaFold, the protein shape prediction software it built. Since AlphaFold’s predictions were made freely available to researchers in 2020, they have been enthusiastically adopted. CEO Demis Hassabis admits bias, thinks AlphaFold is ‘hands down the biggest’ beneficial effects Artificial Intelligence in the World So Far”.

Big Pharma bosses hope these tools will boost their meager productivity.The development cost is estimated Deloitte values ​​each drug at $2.3 billion, with a measly 1.2% return on R&D investment. Artificial intelligence is supposed to improve the problem by accelerating drug discovery. Morgan Stanley says artificial intelligence could cut preclinical development costs by up to two-fifths. It could create a $50 billion market over the next decade.

According to CB Insights, more than 200 new startups are vying for market share. DeepMind’s own drug discovery startup Isomorphic Labs is one of them. Despite the slowdown in venture capital, there has been a flurry of deals recently. Germany’s BioNTech recently acquired Britain’s InstaDeep for $682 million. Eli Lilly signed a $250 million deal with Shenzhen-based Xtalpi in May. In July, Nvidia invested $50 million in US-based Recursion.

Using artificial intelligence does not guarantee success. Oxford-based Exscientia and Sumitomo Pharmaceuticals announced in 2020 that the first clinical trial of an AI-designed molecule had not been successful. In May, London-based BenevolentAI announced it would lay off 180 employees after its lead drug candidate failed.

Nonetheless, Big Pharma’s determination to tap its potential is encouraging. Given that nine out of 10 new drugs fail, there is huge room for improvement.