It’s becoming clear that scientists are only scratching the surface of what artificial intelligence (AI) can teach us about human biology and disease. Case in point: new work by Alphabet’s AI subsidiary, Google DeepMind, that promises to help winnow down the genetic causes of human afflictions.
Recall that it has only been a year since DeepMind researchers rocked the scientific world by unleashing a database with the predicted structure of more than 200 million proteins—like a family album of nearly every protein found in every organism on Earth. Now, it has built on it to develop a tool that can pluck harmful genome mutations from thousands of tiny benign ones.
Called AlphaMissense, the tool is not a quantum leap, but it’s a harbinger for other advances that could spill out of AlphaFold and other such efforts. Last year, Google DeepMind showed that its AlphaFold could predict the structure of any protein from its genetic sequence.
It was hailed as one of the most important scientific advances of the year, albeit with some caveats. Researchers have plumbed it to inform drug-discovery efforts, improve vaccine and drug design, and fill in some of the many blanks about the protein universe.
The magnitude of the advance might not have registered with people, DeepMind’s CEO Demis Hassabis told The Verge, but the ability to translate a string of letters into a complex 3D form has had “the most unequivocally biggest beneficial effects so far in AI on the world." He noted that, “Every Big Pharma company is using it to advance their drug discovery programmes." The tool is free. Now, DeepMind has tweaked AlphaFold to analyse those sequences for an entirely different task: identifying which single-letter changes in the genome might
. Read more on livemint.com