The human brain exhibits great variability between regions at molecular, anatomic and physiological levels. Although this diversity has been known for years, its impact on the neuronal dynamics involving multiple brain regions is still poorly understood.
To address this question, the research teams developed a computational model of the human cerebral cortex that integrates two types of information. On the one hand, structural connectivity, which describes the “wiring” that connects the different regions of the brain. On the other hand, the regional distribution of muscarinic receptors related to neuronal excitability, derived from transcriptomic data and validated with positron emission tomography (PET) scans.
Using this model, the team simulated different brain states, from patterns similar to wakefulness to dynamics characteristic of sleep. The findings showed that the spatial distribution of these receptors contributes to organizing how different brain regions coordinate and how information is transmitted within the neural network.
“Our findings indicate that molecular heterogeneity is not just biological noise,” explains Leonardo Dalla Porta, the article’s first and corresponding author and postdoctoral researcher at IDIBAPS. “The way the wiring and the different receptors are distributed in the brain is fundamental to generating efficient communication patterns and to facilitating transitions between different brain states.”
The study also showed that this spatial organization favours the flow of information between regions and contributes to the emergence of phenomena such as localized slow brain waves, observed both during sleep and in situations associated with lapses in attention or brain lesions.
According to Mavi Sanchez-Vives, head of the IDIBAPS research group, “This work shows that one of the challenges in understanding brain function is integration at different scales, from the molecular information that characterizes each brain region to the macroscopic functional and anatomical connectivity that we can measure in the clinic. Computational models of the brain are fundamental to achieving this.”
Although the findings do not have an immediate clinical application, the authors highlight that this model offers a new conceptual framework for better understanding phenomena observed in patients, such as cognitive deficits after a brain lesion or alterations in consciousness. Moreover, it represents an important step towards the development of personalized computational models that integrate both the connectome and the neurochemical profile of each patient.
Study of reference
L. Dalla Porta, J. Fousek, A. Destexhe & M.V. Sanchez-Vives, Spatially structured heterogeneity shapes large-scale cortical dynamics in a model of the human cortex, Proc. Natl. Acad. Sci. U.S.A. 123 (28) e2532072123, https://doi.org/10.1073/pnas.2532072123 (2026).
