Google’s DeepMind has introduced the latest version of its AlphaFold artificial intelligence model, capable of predicting the structures of various biological molecules.
- The latest model can generate predictions for almost all molecules in the Protein Data Bank (PDB), achieving atomic accuracy.
- Predicted structures include ligands, nucleic acids, and molecules with post-translational modifications (PTMs).
Google‘s DeepMind has unveiled the latest iteration of its AlphaFold artificial intelligence model for drug discovery and biological research.
The new AlphaFold model can predict the 3D structures of various biological molecules, including proteins, ligands, nucleic acids, and post-translational modifications.
Nearly five years ago, DeepMind introduced the original AlphaFold, which had the remarkable ability to accurately predict the structures of various proteins within the human body. Fast forward to 2020, the AI company followed up with the new and improved AlphaFold2 (AF2). This brings us to today. According to their blogpost, the lates model “can now generate predictions for nearly all molecules in the Protein Data Bank (PDB), frequently reaching atomic accuracy.”
The model can predict for:
- Ligands: Small molecules that bind to proteins and other biomolecules.
- Nucleic Acids: Biological molecules that store and transmit genetic information, such as deoxyribonucleic acid (DNA) and ribonucleic acid (RNA).
- Molecules with post-translational modifications (PTMs): protein that has been modified after it has been translated from RNA.
This ability to predict protein-ligand structures is particularly promising for drug discovery. It can help scientists identify and design new molecules that could serve as drugs. This AI has a neat party trick. It can predict these structures without relying on pre-existing information about the protein’s structure or the exact position where the ligand binds to the protein. A far cry from the current docking methods that require some degree of prior knowledge of the structure. I can’t even solve a crossword puzzle without at least one hint…
This means that scientists can explore new proteins that are yet to be characterized. Also, it can represent the inherent flexibility of proteins and nucleic acids as they interact with other molecules. And that is something traditional docking methods fall short of doing.
AlphaFold’s impact has already been felt in the scientific community, with more than 1.4 million users from over 190 countries accessing the AlphaFold Protein Structure Database. Scientists have utilized AlphaFold’s predictions to accelerate research on a wide range of topics, including malaria vaccines, cancer drug development, and environmental solutions.
DeepMind’s AI can come in handy when it comes to disease pathways, genomics, biorenewable materials, plant immunity, and potential therapeutic targets.
Oh, the places you will go, AlphaFold!
Inside Telecom provides you with an extensive list of content covering all aspects of the tech industry. Keep an eye on our Intelligent Tech sections to stay informed and up-to-date with our daily articles.