When molecules react: New software visualizes biological processes in motion
From energy production to genetic changes: molecules are constantly in motion within biological cells. However, experimentally investigating such processes, which occur on very small length and time scales, is extremely difficult. To overcome these challenges, a research team at the Max Planck Institute for Polymer Research led by Director Frauke Gräter has now developed a new simulation method. This method works extremely quickly and can predict chemical processes in cells with high precision.
Chemical reactions drive life. They ensure that cells obtain energy, proteins perform their functions, and DNA changes under certain conditions. However, many of these processes occur on extremely small scales - so small and so fast that they are difficult to observe directly through experiments.
Researchers have therefore been using computer simulations for years to study the behavior of molecules. However, one important property has mostly been left out so far in order to keep the simulations within the limits of what is feasible on today’s supercomputers: While molecules move realistically in many simulations, chemical bonds cannot break or form anew.
A research team at the Max Planck Institute for Polymer Research (MPI-P) in Mainz has now developed a method to overcome this limitation. The new software KIMMDY (short for KInetic Monte Carlo MolecularDYnamics) combines various computational approaches and uses machine learning methods to calculate when and where chemical reactions can occur.
“This allows us not only to track how molecules move, but also how they react with one another,” says Professor Frauke Gräter, director of the “Biomolecular Mechanics” department at MPI-P. “This, in turn, opens up entirely new possibilities for investigating complex biological processes on a computer.”
The newly developed method makes it possible to simulate very large molecular systems—such as proteins or DNA in their natural environment—while also tracking reaction chains in which one chemical step triggers the next. Such processes play a role in many biological contexts, such as damage to biomolecules or chemical modifications within proteins or DNA.
To demonstrate the method’s capabilities, the researchers examined several examples from biology. In simulations of collagen, a protein crucial for the stability of our skin, bones, and connective tissue, they were able to track how reactive molecular fragments migrate through the protein and accumulate at specific sites. Damage to DNA, such as that caused by UV radiation, can now also be investigated.
The new method stands out because it allows systems with millions of atoms to be calculated more efficiently than in competing approaches. This means that KIMMDY could help us better understand biological and chemical processes in the future. At the same time, KIMMDY opens up new possibilities for interpreting experimental results and planning new experiments.
The researchers from the Max Planck Institute for Polymer Research recently published their findings in the prestigious journal Nature Communications.
Wissenschaftlicher Ansprechpartner:
Prof. Frauke Gräter
graeter@mpip-mainz.mpg.de
Originalpublikation:
Hartmann, E.; Buhr, J.; Riedmiller, K.; Ulanov, E.; Schüpp, B.; Kiesewetter, D.; Sucerquia, D.; Aponte-Santamaría, C.; Gräter, F.
KIMMDY: a biomolecular reaction emulator
Nat Commun 17, 3500 (2026)
https://dx.doi.org/10.1038/s41467-026-71955-2
Ähnliche Pressemitteilungen im idw
