A new era of biological research has opened with artificial intelligence (AI) predicting the 3D shape of nearly every protein known to science — just a year after its first data were released. Thanks to AlphaFold, an AI tool developed by Google-owned AI company DeepMind, more than 200 million protein structures are now shared online in a freely accessible and searchable database called AlphaFold DB.
This achievement paved the way for countless scientific discoveries in proteins, the building blocks of life. The researchers were giddy with excitement.
“Determining the 3D structure of a protein that once took months or years now takes seconds,” cardiologist Eric Topol of the Scripps Research Translational Institute said in a statement about the data release. Scientist, Tuesday (2/8/2022).
“With the addition of new structures that illuminate almost the entire protein universe, we can expect that more biological mysteries will be solved every day.”
In collaboration with scientists from the European Molecular Biology Laboratory’s European Bioinformatics Institute (EMBL-EBI), DeepMind launched the first batch of AlphaFold predictions in July last year.
Declared as a revolutionary tool that will transform biological research and accelerate drug discovery, AlphaFold predicts the 3D shape of proteins based on their amino acid sequences. Linked together in chains, this sequence of amino acids assembles long proteins that are folded into pleated sheets and twisted into bands.
By understanding the folding shape of a particular protein, scientists can better understand how the protein works, by understanding its main role in cells. AlphaFold is designed to accelerate that process, with this latest data release predicting more than 200 million protein structures found in plants, bacteria, animals and other organisms.
“This wish has come true sooner than we dared to dream,” DeepMind Chief Executive Demis Hasabis said in a statement regarding the latest data release.
Already, researchers are using the first batch of AlphaFold’s predictions to improve their understanding of deadly diseases like malaria, open the door to better vaccines, and solve biological puzzles about a giant protein that has baffled scientists for decades. is used. Not to mention identifying a never-before-seen enzyme that could help recycle plastic pollution.
“Alphafold has sent ripples through the molecular biology community,” said Sameer Velankar, lead structural biologist at the EMBL-EBI Protein Data Bank.
“In the past year alone, there have been over a thousand scientific articles on various research topics using the alphafold structure; I’ve never seen anything like it. And that’s just the effect of a million predictions,” Velankar said.
“Imagine the impact of more than 200 million publicly accessible protein structure predictions in the AlphaFold database.”