ABOUT US
Empowering Science with Precision and Simplicity!
Why Did the Nobel Prize in Chemistry Go to AI Again? Let AI Empower Your Proteomics Research!
Release time:
2025-01-09
Two of this year's three major Nobel Prizes in Science were awarded to AI, making it clear that AI has become a significant force driving technological advancement.
This year, the Nobel Prize in Chemistry was split into two halves, awarded to two outstanding scientists and their teams. One half went to Professor David Baker from the University of Washington, Seattle, for his remarkable contributions to the field of "computational protein design."
David Baker’s innovative computational methods successfully designed proteins with specific functions. This achievement not only advanced protein science but also opened new possibilities for biotechnological and medical applications.
The other half of the award went to Demis Hassabis and John M. Jumper from Google DeepMind, London, for their groundbreaking work in "protein structure prediction."
Google DeepMind leveraged advanced artificial intelligence technologies to significantly improve the accuracy and efficiency of protein structure predictions. This provided powerful tools for understanding protein functions, disease mechanisms, and drug development.
The Nobel Prize committee’s official evaluation described proteins as "ingenious chemical tools of life." It recognized that David Baker had accomplished the nearly impossible feat of building entirely new types of proteins, while Demis Hassabis and John Jumper had developed an AI model that realized a dream spanning over 50 years: predicting protein structures from their amino acid sequences.
AlphaFold's Development
In proteins, amino acids are linked in long chains, which then fold into three-dimensional structures. These structures are critical for the protein's function. Since the 1970s, researchers have been trying to predict protein structures from their amino acid sequences.
However, the breakthrough in achieving accurate and efficient protein structure prediction actually started with the game of Go. In 2016, AlphaGo defeated South Korean professional Go player Lee Sedol 4-1. The updated version of AlphaGo then beat the world’s top-ranked Go player, Ke Jie, 3-0.
At that time, no one imagined that AlphaGo’s impact would extend beyond the field of Go. Soon after, the AlphaGo team at DeepMind shifted its focus to the field of biological sciences.
In January 2020, the AlphaFold1 paper was published in Nature, demonstrating a level of protein structure prediction accuracy far beyond expectations.
AlphaFold2：
In 2020, Demis Hassabis and John Jumper introduced an AI model called AlphaFold2.
With its help, researchers have identified the structures of nearly 200 million proteins. Since 2020, AlphaFold2 has been used by over 2 million researchers from 190 countries/regions, and the paper on AlphaFold2 was published in July 2021.
However, until the AlphaFold2 model, AlphaFold was able to predict the structures of single-chain proteins, but it couldn’t predict the structures of more complex and biologically significant macromolecular complexes.
AlphaFold-Multimer:
The introduction of AlphaFold-Multimer enabled the prediction of protein-protein complex structures.
AlphaFold3：
In 2023, AlphaFold3 was published in Nature, marking a major milestone where the AlphaFold model expanded to predict not only protein-protein complex structures but also the structures of proteins, DNA, RNA, and small molecular ligands.
Related News
2025-07-10
2025-07-08
2025-07-03
Literature Sharing | Regulation of co-translational mRNA decay by PAP and DXO1 in Arabidopsis
2025-07-01
2025-06-27
2025-06-24
2025-06-20
Contact Us
Tel: +86 025-85205672
Email: info@pronetbio.com
Address: Building 3C, Nanjing Xianlin Zhigu,
Qixia District, Nanjing, Jiangsu, China, 210033
Need more info?
Let's connect!