Artificial intelligence company DeepMind has announced a major medical science breakthrough in determining the structure of nearly 200 million proteins.
Proteins are not two-dimensional molecules, but have chemical properties determined by their three-dimensional shape – but finding these shapes is an intensive process.
The breakthrough has significant implications for medicine, and new Google-backed DeepMind research has been hailed as having “the potential to dramatically increase our understanding of biology”.
A protein is made of a sequence of amino acids, but without knowing how these chains are connected it is impossible to know how they interact with human cells and can be modified.
Last year, DeepMind, owned by Google’s parent company Alphabet, shared the results of an AI system called AlphaFold can predict the 3D structure of a protein from its unidimensional amino acid sequence.
A year before that, PC gamers were asked donate some of their computing power for an international effort to study diseases including COVID-19 and Alzheimer’s disease to simulate the molecular dynamics of protein folding.
It’s an important topic for medical science because the structure of proteins determines the chemical reactions in human cells, and in general, the human body as a whole – but so far, only a portion. small known protein structure.
The announcement of DeepMind and its protein structure database – which is being freely shared – dramatically increased the number of known protein structures from nearly a million to over 200 million.
It was created in conjunction with EMBL’s European Bioinformatics Institute (EMBL-EBI) which director general Edith Heard said: “AlphaFold now offers a 3D view of the protein universe.”
“We were amazed at the speed with which AlphaFold has become an essential tool for hundreds of thousands of scientists in laboratories and schools,” said Demis Hassabis, DeepMind founder and CEO. universities worldwide.
Mr. Hassabis added: “From fighting disease to tackling plastic pollution, AlphaFold has made an incredible impact on some of our biggest global challenges.
“We hope that this expanded database will further support countless more scientists in their important work and open up entirely new avenues of scientific discovery.”
The research has been lauded by scientists who have used the AlphaFold model to develop malaria antibodies and even special enzymes that can break down plastic.
More than 1,000 scientific articles have been published since its launch, and more than 500,000 researchers from over 190 countries have accessed the database.
Additional areas of research supported by the database include honey bee health, understanding how ice forms, and neglected diseases such as Chugs disease and Leishmaniasis.
“This is just the impact of a million predictions; imagine the impact of having over 200 million protein structure predictions publicly accessible in the AlphaFold Database,” said Sameer Velankar, head of the banking group. EMBL-EBI’s European protein data row.