Researchers at Arizona State University announced on Mar. 27 a new method to uncover slow, hidden movements in proteins, which could help improve drug design and understanding of diseases. The team, led by associate professor Matthias Heyden from ASU's School of Molecular Sciences, published their findings in Science Advances.
The study is significant because proteins are essential biomolecules with complex structures that perform many functions in living cells. Their ability to change shape influences processes such as tissue repair, metabolism, and immune response. Understanding how proteins move can help scientists predict their behavior and create more effective drugs.
Heyden said his group developed a technique that identifies low-frequency vibrations in proteins using short computer simulations lasting only billionths of a second. "In short, we resurrected a longstanding idea that conformational transitions in proteins are tied to low-frequency vibrations," Heyden said about the approach.
He explained further: "We developed a method to identify these vibrations through natural fluctuations caused by molecular collisions. The natural motions stand out if analyzed with the right tools." He compared this process to checking if an unlocked door opens easily without needing to force it off its hinges: "On a molecular scale, it is even enough to observe tiny fluctuations that are always present at room temperature." Knowing these vibrations should allow researchers "to speed up the sampling of conformational transitions in molecular dynamics simulations," Heyden said.
The research used powerful graphics processors on ASU's supercomputer "Sol" and allowed scientists to watch meaningful protein shape changes within less than a day—a process that previously took weeks or months. This rapid simulation capability could lead to designing more dynamic proteins for medical applications and better understanding allosteric effects—where changes at one site affect distant parts of the molecule—potentially resulting in drugs with fewer side effects.
Heyden also noted the relevance for artificial intelligence-driven protein studies: fast simulation methods like theirs may expand current approaches from predicting static structures based on sequence alone toward capturing full dynamics as well.
The work received support from the National Science Foundation (CHE-2154834) and the National Institutes of Health (R01GM148622).
