If a specific sequence of amino acids is known, biochemists reasoned that it could be possible to predict the complicated 3D structure that the sequence folds into. Since 1994, protein researchers have held a biennial challenge—the Critical Assessment of Structure Prediction (CASP) experiment—to test theoretical structure predictions against experimental observations. AI models are adept at pattern recognition and have been steadily improving the prediction accuracy for years. As recounted in Physics Today's 2021 report on DeepMind's transformative work, designed by Hassabis, Jumper, and colleagues:
Structure predictions are graded on a scale from 0 to 100: A random guess might score below 20; an atomically precise structure, above 90.
From the early days of CASP, models have been scoring above 80 for the easiest template-based predictions, while scores for the most difficult targets have been stuck around 40. So DeepMind's first CASP entry, the original AlphaFold model in 2018, shook things up by scoring above 70 for more than half of the most difficult targets. …
For the 2020 CASP assessment, the DeepMind team had revamped its model into AlphaFold2, whose predictions scored near 90 even for the most difficult targets—scores so high that they were probably limited by the imprecision of the experimental structures the predictions were graded against.
Earlier this year, Hassabis, Jumper, and colleagues reported the development of AlphaFold3, an upgraded model that also predicts the structures of nucleic acids, small molecules, and other biochemical complexes. The accurate design of proteins and the prediction of their structure have far-reaching implications in medicine and many other fields.
At the Nobel press conference, Baker offered an example: an inexpensive nasal spray that he and colleagues are working on to protect against multiple coronavirus variants. "I'm really excited," he said, "about all the ways in which protein design can make the world a better place."
No comments:
Post a Comment