Home Technology

Alphabet’s new company is using artificial intelligence for drug research

Alphabet’s new company will use artificial intelligence methods in drug discovery, Google’s parent company announced. The company builds on the work of DeepMind, another subsidiary of Alphabet, which has pioneered the prediction of protein structure using artificial intelligence.

A new company called Isomorphic Laboratories will use this success to develop tools that can help identify new drugs. Demis Hassabis, CEO of DeepMind, will also be CEO of Isomorphic, but the two companies will remain separate and only work together occasionally.

Experts have been waiting for years for artificial intelligence to make new drugs faster and cheaper to find a variety of diseases to treat. Artificial intelligence can help you search databases of potential molecules to find, for example, the ones that best fit a particular biological target, but technology can play a role in fine-tuning proposed compounds. Hundreds of millions of dollars have been invested in companies developing artificial intelligence devices over the past two years.

Isomorphic will try to create models that can predict how drugs interact with the organization, Hassabis said. To Stat News. Work on the protein structure of DeepMind can help you figure out how multiple proteins can interact with each other. The company may not be developing its own drugs, but will rather sell its models. It will focus on building partnerships with pharmaceutical companies, a company spokesman told The Verge

However, developing and testing drugs can be more challenging than exploring the structure of proteins. For example, even if two proteins have a physically compatible structure, it is difficult to say how much they will actually fit together. A drug candidate that looks promising based on its chemical level function may not always work when administered to a living organism. More than 90 percent of the drugs that reach the clinical trial end up failing, as chemist and writer Derek Lowe pointed out in the journal Science. Most of the problems are not because there was something wrong at the molecular level.

The work at DeepMind and the work proposed at Isomorphic may help break through some research bottlenecks, but they do not provide a quick solution to the myriad challenges of drug development. “Laborious, resource-intensive work, such as performing biochemical and biological evaluations of drug functions,” remains, as Helen Walden, professor of structural biology at the University of Glasgow, told The Verge.

Hardware, software, tests, curiosities and colorful news from the world of IT by clicking here

NO COMMENTS

Leave a ReplyCancel reply

Exit mobile version