DeepMind’s AlphaFold Discovers Nearly Every Known Protein
AI and Data Science NewsAlphaFolddeepmindposted by ODSC Team August 3, 2022 ODSC Team
In a stunning development, Google’s DeepMind AI has discovered nearly every single protein currently known to science. This was achieved thanks to using the AlphaFold program which was first developed back in 2018, and then released to the public last summer. The open-source program uses machine learning algorithms to predict a protein’s three-dimensional structure using its sequence of amino acids.
Amino acids are the building blocks of proteins that link together in rows of chains. DeepMind was able to use that information to put together an array of proteins. As it stands, the database currently holds over 200 million protein structures that were identified by AlphaFold. These same proteins are found in all known bacteria, plants, animals, and other organisms identified thus far. This growing database can potentially give researchers a leg up on new protein workhorses that can provide a variety of benefits.
As Cardioiologist Eric Topol from the Scripps Research Translational Institute explains in a statement about the news, “Determining the 3D structure of a protein used to take many months or years, it now takes seconds…With this new addition of structures illuminating nearly the entire protein universe, we can expect more biological mysteries to be solved each day.”
This could have major implications in the world of research where costs often rise due to the time it takes to discover and understand protein structures. So not only could this be a time saver when it comes to scientific research, but free up resources that are often extremely finite.
DeepMind’s Chief Executive, Demis Hassabis said, “That hope has become a reality far quicker than we had dared to dream.” He goes on further in a series of Tweets showcasing the breakthrough.
Part of what DeepMind has already done is assisting researchers to characterize a key malaria parasite protein. Before DeepMind, it had not been amenable to X-Ray crystallography. In the future, this can positively assist in future vaccine development against malaria.
How this will affect the microbiology community, medical community and many others is impossible to predict. But one likely outcome will be that discovery of new duets will accelerate thanks to Deep Mind and the power of artificial intelligence.