Isomorphic Labs, a spinoff of Google and DeepMind, has developed a new AI model that reportedly is able to predict the structures and interactions of every life’s molecule with greater accuracy than current tools.
DeepMind’s co-founder Demis Hassabis states that AlphaFold 3 “can predict the structures and interactions of nearly all of life’s molecules with state-of-the-art accuracy including proteins, DNA and RNA.”
The interactions between proteins, which can range from enzymes vital to human metabolism to antibodies that combat infectious diseases, are crucial for drug discovery and development.
DeepMind announced that the results, which were published in the research journal Nature on May 8, could significantly decrease the time and cost required to develop treatments that may change lives.
“With these new capabilities, we can design a molecule that will bind to a specific place on a protein, and we can predict how strongly it will bind,” Hassabis said in a press statement.
We're excited to announce #AlphaFold 3 with @GoogleDeepMind in @Nature: our new AI model for predicting biomolecule structures with unprecedented breadth and accuracy.
Expanding beyond proteins to tackle DNA, RNA, small molecules to fuel advances in biology & drug design ???? pic.twitter.com/s5OdfGMCG7
— Isomorphic Labs (@IsomorphicLabs) May 8, 2024
AlphaFold has previously transformed research by simplifying the prediction of protein 3D structures. However, in the past, it could not predict scenarios where a protein binds with another molecule, which has now been improved with AlphaFold 3.
Although restricted to non-commercial use, researchers are said to be enthusiastic about its expanded predictive capabilities and the potential for accelerating drug discovery.
In a statement on X, Isomorphic Labs said: “With AlphaFold 3 we can now create and test hypotheses at the atomic level, and produce highly accurate structure predictions within seconds.”
It then added in a blog post: “This breakthrough opens up exciting possibilities for drug discovery, allowing us to rationally develop therapeutics against targets that were previously difficult or deemed intractable to modulate.”
How could #AlphaFold 3 transform drug discovery?
Most drugs are small molecules known as ligands that bind to proteins to change how they interact in human health and disease.
AlphaFold 3 can predict these interactions to atomic accuracy. pic.twitter.com/TmefhHChmm
— Isomorphic Labs (@IsomorphicLabs) May 8, 2024
How to access AlphaFold Server
Google DeepMind and Isomorphic Labs have announced that scientists will have free access to most features of their newly launched research tool, the AlphaFold Server.
Announcing AlphaFold 3: our state-of-the-art AI model for predicting the structure and interactions of all life’s molecules. ????
Here’s how we built it with @IsomorphicLabs and what it means for biology. ???? https://t.co/K7uxMxdNr8 pic.twitter.com/MJfJ6NCEFW
— Google DeepMind (@GoogleDeepMind) May 8, 2024
To use AlphaFold 3’s potential in drug design, Isomorphic Labs is reportedly partnering with pharmaceutical companies to address real-world drug design challenges and ultimately develop new transformative treatments for patients.
Since 2021, non-commercial researchers have had free access to AlphaFold’s predictions through a database that includes over 200 million protein structures. This resource has been cited thousands of times in scholarly work.
DeepMind announced that the new server simplifies the process, enabling researchers to conduct experiments with just a few clicks.
AlphaFold Server will allow inputting data through its web interface for assorted biological molecule types, using a FASTA file. The AI model then processes the job and returns an overview of the structure as a 3D rendering.
Featured image: Google DeepMind
