Nearly every protein that is known to exist on Earth has had its structure predicted using the innovative artificial intelligence (AI) network AlphaFold, which has been used by researchers.
The European Bioinformatics Institute (EMBL-EBI), a non-profit organisation located close to Cambridge, UK, and DeepMind, a London-based artificial intelligence startup that developed AlphaFold and is owned by Google, together constructed a database where the data dump is freely available.
Demis Hassabis, CEO of DeepMind, said during a press conference that “basically, you can think of it as spanning the entire protein universe.” In the field of digital biology, a new age has begun.
The three-dimensional shape or structure of a protein determines its biological function. The creation of precise maps of the amino-acid configuration of proteins is typically the first step in learning how proteins work, as the majority of drugs are generated using structural data.
The AlphaFold network was developed by DeepMind using deep learning, a type of artificial intelligence, and the AlphaFold database, which contains more than 350,000 structure predictions covering nearly every protein produced by people, mice, and 19 other organisms with extensive research, was made public a year ago. There are currently close to one million entries in the catalogue.
We’re getting ready to reveal this vast resource, says Christine Orengo, a computational biologist at University College London who has used the AlphaFold database to find novel protein families. Having all the information predicted for us is excellent.