For decades, scientists have been trying to figure out how to quickly predict the twisted, tangled shapes of proteins, and from there, unlock a deeper understanding of the mechanisms of life itself.
This week, an artificial intelligence program created by Google’s sister company, DeepMind, proved to actually overcome this challenge. It predicted the way in which proteins were twisted into three-dimensional structures in the biennial competition. The judgement is considered to be a change. The result of the rules of the game.
John Moult, a computational biologist at the University of Maryland, said: “In a sense, the problem has been solved.” He co-authored “Key Evaluation Techniques for Protein Structure Prediction Technologies” in 1994. (CASP) competition and was cited by the journal Nature.
Janet Thornton of the European Institute of Bioinformatics said in a statement to the organizers of the competition on Monday that the problem-solving work is “a victory of human curiosity, effort and wisdom.”
Thornton, who has nothing to do with CASP or DeepMind, said: “A better understanding of protein structure and the ability to use computers to make predictions means a better understanding of life, evolution, and human health and disease.”
The cells of all living things contain thousands of proteins, which are the main force that catalyzes most chemical reactions in the body.
They are essential to life. From muscle function to oxygen carrying in the blood, they are also the key to diseases such as cancer and even COVID-19.
Proteins start with amino acid chains and then crimp into unique three-dimensional tangles.
It is this shape that is directly related to its function.
For half a century, scientists have been thinking about how to accurately predict the formation process of protein amino acids, and this possibility may take several years, and this possibility is produced by the amino acid chain of the protein, and the number is so large. Unbelievable.
The CASP competition involves about 100 teams. These teams have obtained the amino acid sequences of dozens of proteins and are responsible for estimating their final shape, which is known to the organizers.
DeepMind’s program AlphaGo defeated elite human players in the complex strategy game “Go” and shocked the world in 2016. In the last competition, DeepMind has been among the best.
According to the organizers of CASP, the AlphaFold program determined the shape of many proteins with “accuracy comparable to that achieved by expensive and time-consuming laboratory experiments.”
“This changed medicine”
Andrei Lupas, an evolutionary biologist at the Max Planck Institute for Developmental Biology, who is a member of the evaluation team, told Nature that AlphaFold helped him determine a kind of his experiment Chamber has been trying to suppress the structure of the protein.
He told Nature: “This will change medicine. It will change research. It will change bioengineering. It will change everything.”
Derek Lowe, of an article on drug discovery and the pharmaceutical industry in Science Translational Medicine, described protein folding as “watching piles of hinged wood spontaneously restacking into functional boats, Truck and tree house”.
He said that AlphaFold results do not mean that the program will continue to provide the correct protein structure.
“But to achieve such structural accuracy on many different proteins, this is something that has never been done before.”
DeepMind said it is studying how the program can help increase understanding of certain diseases, such as determining whether a protein is malfunctioning.
It said in a statement: “These insights can make drug development work more precise, thereby supplementing existing experimental methods and finding promising treatments faster.”
The company added that it is working on a peer-reviewed paper and is “exploring how to best provide broader access to the system.”
The iPhone 12 Pro series is amazing, but why is it so expensive in India? We discussed it on the weekly technical podcast Orbital, you can subscribe via Apple Podcast, Google Podcast or RSS, download the episode, or click the play button below.