DeepMind breakthrough as AI firm unlocks 3D structure of all known proteins

AI research lab DeepMind has announced that it has cracked the structure of all known proteins, thanks to advances in the Alphabet-owned company’s AlphaFold Software.

It used to take scientists months or years to understand a proteins structure, and DeepMind (formerly known as Google DeepMind) stunned many in the scientific community 12 months ago with the release of predicted structures for thousands of proteins.

Now, in a blog published last week, DeepMind revealed that its catalogue of structures has expanded 200 times with the release of over 200 million protein structures from bacteria to humans.

AlphaFold’s newly predicted structures were released last week to an existing database through a partnership with the European Molecular Biology Laboratory’s European Bioinformatics Institute (EMBL-EBI).

The AlphaFold DB now means that complex information is only a google search away – with DeepMind reporting that it takes approximately “10-20 seconds to make each protein prediction”.

Demis Hassabis, founder and CEO of DeepMind, hailed the breakthrough as a “new era in digital biology” in which drug developers could now examine AI-predicted structures of proteins to design small molecules that influence those proteins and therefore treat illness.

Hassabis added that the findings have opened many new opportunities for researchers to use AlphaFold to advance their work on important issues, including sustainability, food insecurity, and neglected diseases.

According to DeepMind, in just twelve months, AlphaFold has been accessed by more than half a million researchers and used to accelerate progress on important real-world problems ranging from plastic pollution to antibiotic resistance.

Across other sectors, DeepMind has also been helping historians and archaeologists develop new tools to help them interpret and restore the text on damaged ancient Greek inscriptions.

Hassabis added that DeepMind was also expanding its AI for Science team to accelerate further progress on ”fundamental biology research” and apply AI to other important scientific challenges, such as climate science, quantum chemistry and fusion.