An artificial intelligence group says its program has successfully predicted the structure of nearly every protein known to science — effectively solving one of biology’s “great challenges” and paving the way for new discoveries and technologies in fields as diverse as medicine, food security and climate science.
DeepMind, the AI company owned by Google’s parent company, Alphabet, announced Thursday that the AlphaFold program has expanded its open online database to include more than 200 million protein structures.
Demis Hassabis, CEO of DeepMind, said at a news briefing that the extensive catalog now includes “the entire protein world,” from the genome sequences of nearly every organism on the planet.
Proteins are long, complex chains of amino acids that are the building blocks of life. Scientists have long sought to untangle how these chains elegantly bend and fold into 3D shapes because understanding their structure can yield valuable insights into their function. Knowing the specific shape of a protein and how its various molecules interact, for example, can help researchers narrow down potential targets for medical treatments.
The developed AlphaFold database includes protein structures of plants, bacteria, animals and other organisms, according to DeepMind.
These updates provide “new opportunities for researchers to use AlphaFold to advance their work on important issues, including sustainability, food insecurity and neglected diseases,” Hasabis wrote in a blog post published Thursday about the achievement.
“By demonstrating that AI can accurately predict the shape of a protein down to atomic precision, at scales and in minutes, AlphaFold has not only provided a solution to a great 50-year challenge, but has also become the first major proof of our foundational thesis: that AI can accelerate scientific discovery in a significant way, thus advancing humanity forward.”
AlphaFold was introduced in 2020, and last year DeepMind dazzled the scientific community by revealing a catalog of structures that contained nearly every protein in the human body. The AlphaFold protein structure database, built in collaboration with the European Molecular Biology Laboratory, contained hundreds of thousands of newly predicted protein structures.
Researchers around the world are already using the vast amount of information to study topics ranging from antibiotic resistance to plastic pollution, according to Hasabis.
Researchers at the University of Portsmouth in the UK, for example, announced in July 2021 that they were using the database to help engineer enzymes to recycle certain types of single-use plastics.
“AlphaFold provides us with an exciting new library of templates to engineer faster, more stable and cheaper enzymes to recycle plastic,” John McGeehan, director of the University of Portsmouth’s Center for Enzyme Innovation, said in a statement at the time.
Hasabis said DeepMind is expanding its database, with a particular focus on applications related to drug development, basic biology research, climate science, quantum chemistry, and fusion.
“AlphaFold is a glimpse into the future, and what might be possible with computational methods and artificial intelligence applied to biology,” he wrote.
#Protein #predictions #group #solved #biologys #big #challenges