Google DeepMind has just released high-quality AI predictions for the shape of every single protein in the human body, as well as for the proteins of 20 additional organisms that scientists rely on for their research. How so? Researchers used AlphaFold 2 to predict 350,000 protein structures belonging to humans as well as other organisms, and this could accelerate the discovery of new drugs to treat disease, alongside other applications. Read more for three videos and additional information.
In the nuclei of human cells are the instructions for making human proteins are contained in our genomes, and there are about 20,000 of these proteins expressed by the human genome – biologists refer to this full complement as the “proteome”. AlphaFold 2 was able to make a confident prediction of the structural positions for 58% of the amino acids, with 35.7% of them predicted with a very high degree of confidence DeepMind aims to greatly expand the coverage in the database to almost every sequenced protein known to science, or over 100 million structures.
- System: AMD Ryzen 3 3100 3.6GHz 4-Core | AMD A520 Chipset | 8GB DDR4 | 240GB SSD | 2TB HDD | Genuine Windows 10 Home 64-bit
- Graphics: AMD Radeon RX 550 2GB Video Card | 1x HDMI | 1x DisplayPort
- Connectivity: 6 x USB 3.1 | 2 x USB 2.0 | 1x RJ-45 Network Ethernet 10/100/1000 | 802.11AC Wi-Fi | Audio: 7.1 Channel | Keyboard and mouse
Determining protein structures experimentally is a time-consuming and painstaking pursuit, but AlphaFold demonstrated that AI could accurately predict the shape of a protein, at scale and in minutes, down to atomic accuracy,” said the DeepMind team.