Multiscale simulations successfully connect micro- and macro levels of brain activity

More recently, digital research tools have been developed that can integrate data across scales and connect to each other. Building on these tools, scientists at Paris Saclay University have now introduced a new multiscale modeling approach that can simulate how microscopic changes are translated across multiple levels of brain organisation to impact macroscale brain activity.

The simulations connected single neuron models, spiking neural networks and mean-field models with a whole-brain network simulation. The computational simulations successfully predicted how the effects of anesthesia on synaptic receptors can lead to the transitions of macroscale brain activity that have been observed empirically. The modeling framework has been presented in the journal Nature Computational Science.

A computational method that brigdes scales and provides accurate predictions is of high interest for pharmaceutical treatments for brain diseases: Many pharmacological compounds in drugs act at the molecular level and generate macro-level modifications to brain states. Tracking these mechanisms better could ultimately support more targeted drug design.

For their modeling framework, the team combined a range of digital research technologies that had been developed in the European Flagship Human Brain Project, and which are available on the EBRAINS research infrastructure. The research was supported by the French CNRS and Agence Nationale de La Recherche and by the European Union through the Human Brain Project and the Virtual Brain Twin project.

Find the full open access paper and a summary (“research briefing”) here:

Sacha, M., Tesler, F., Cofre, R. et al. A computational approach to evaluate how molecular mechanisms impact large-scale brain activity. Nat Comput Sci 5, 405–417 (2025). https://doi.org/10.1038/s43588-025-00796-8

Tesler, F. & Destexhe, A.: How molecular changes impact brain states and whole-brain activity: a multiscale approach. Nat Comput Sci 5, 442–443 (2025). https://doi.org/10.1038/s43588-025-00813-w