By Cade Metz
Artificial intelligence gives machines the power to generate videos, write computer code, and even carry on a conversation.
It also accelerates efforts to understand the human body and fight disease.
On Wednesday, Google DeepMind, the tech giant’s core artificial intelligence lab, and Isomorphic Labs, a sister company, unveiled a more powerful version of AlphaFold, an artificial intelligence technology that helps scientists understand behavior microscopic mechanisms that drive the cells of the system. human body.
An early version of AlphaFold, released in 2020, solved a puzzle that had tormented scientists for more than 50 years. This was called “the protein folding problem.”
Proteins are microscopic molecules that determine the behavior of all living things. These molecules start as chains of chemical compounds before twisting and folding into three-dimensional shapes that define how they interact with other microscopic mechanisms in the body.
Biologists have spent years, even decades, trying to identify the shape of individual proteins. Then AlphaFold came along. When a scientist fed this technology with a string of amino acids that make up a protein, he was able to predict the three-dimensional shape within minutes.
When DeepMind released AlphaFold to the public a year later, biologists began using it to accelerate drug discovery. Researchers at the University of California, San Francisco, used the technology to understand the coronavirus and prepare for similar pandemics. Others used it as they struggled to find cures for malaria and Parkinson’s disease.
The hope is that this type of technology will greatly streamline the creation of new drugs and vaccines.
“This tells us a lot more about how the machines in the cell interact,” said Google DeepMind researcher John Jumper. “It tells us how it should work and what happens when we get sick.”
The new version of AlphaFold — AlphaFold3 — extends the technology beyond protein folding. In addition to predicting the shape of proteins, it can predict the behavior of other microscopic biological mechanisms, including DNA, where the organism stores genetic information, and RNA, which transfers information from DNA to proteins .
“Biology is a dynamic system. You need to understand the interactions between different molecules and structures,” said Demis Hassabis, general manager of Google DeepMind and founder of Isomorphic Labs, which Google also owns. “This is a step in that direction.”
The company offers a website where scientists can use AlphaFold3. Other laboratories, notably that of the University of Washington, offer similar technology. In a paper published Tuesday in the scientific journal Nature, Dr. Jumper and his fellow researchers show that it achieves a level of precision well beyond the state of the art.
The technology could “save months of experimental work and enable research that was previously impossible,” said Deniz Kavi, co-founder and chief executive of Tamarind Bio, a startup that develops technology to accelerate drug discovery.
“This represents tremendous promise.”
– It was called “the protein folding problem”
– The new version — AlphaFold3 — predicts protein shapes, in addition to predicting protein shapes
– It can predict the behavior of other microscopic biological mechanisms, including DNA and RNA.
– The company offers a website where scientists can use AlphaFold3
©2024 New York Times News Service
First publication: May 10, 2024 | 01:23 STI
