Follow us on Google News (click on ☆)
Structural biology focuses on understanding the molecular structures of biological materials, such as proteins and nucleic acids. AlphaFold3 goes beyond its predecessors by offering increased accuracy in predicting interactions between proteins and other biomolecules, such as DNA and RNA.
AlphaFold3 enables researchers to better understand the fundamental molecular mechanisms of life. By predicting how proteins bind and interact with other biological molecules, it paves the way for innovations in various fields. For example, in agriculture, AlphaFold3 could help develop more resilient crops. In the medical field, it could accelerate the discovery of new drugs.
A study published in Nature on May 8 highlights that the program can predict interactions not only between large molecules like proteins but also with small molecules called ligands. These ligands play a crucial role by binding to receptors on proteins, much like a key fits into a lock. AlphaFold3 can also model the chemical modifications of these biomolecules, which may be involved in diseases.
Researchers claim that AlphaFold3 is 50% more accurate than current methods in predicting protein structures and their interactions. In pharmaceutical research, AlphaFold3 has outperformed existing docking programs and RoseTTAFold All-Atom, a neural network for predicting biomolecular structures. Frank Uhlmann, a biochemist at the Francis Crick Institute in London, uses AlphaFold3 to predict the structures of proteins interacting with DNA, with mostly accurate results.
However, AlphaFold3 is no longer open source, thus limiting scientists' access to the model's code and training data. For non-commercial research, researchers can use the recently launched AlphaFold server. This free service allows submitting molecular sequences and receiving predictions within minutes, although it is limited to 20 tasks per day.
AlphaFold3 represents a giant leap in structural biology and opens new horizons for scientific and medical research. Its potential to revolutionize the understanding of biomolecular interactions is immense, promising significant breakthroughs in various fields.