New computer model connects brain chemistry to neural activity patterns

Researchers have developed a computer model of the human cortex that links its microscopic chemistry to its brain-wide patterns of activity, providing new evidence that regional differences in receptor density help shape how activity and information move across the brain.

The study was published in the Proceedings of the National Academy of Sciences (PNAS). It was developed using The Virtual Brain (TVB), an open-source whole-brain simulation platform that is part of the EBRAINS research infrastructure, and was supported in part through the EBRAINS 2.0 project and The Virtual Brain Twin Project.

One of the central challenges in neuroscience is understanding how processes occurring at the molecular level influence the behaviour of the brain as a whole. Our study provides a concrete example of how we can begin connecting these very different scales within the same computational framework."

Leonardo Dalla Porta, researcher at the Institute of Biomedical Investigations August Pi i Sunyer (IDIBAPS) and first author of the study

Most large-scale brain models simplify things by treating every cortical region as if it worked the same way. This new model takes a different approach: it incorporates detailed maps of muscarinic acetylcholine receptor density – a key target of the neuromodulator acetylcholine – across 68 brain regions, layered onto the brain's actual structural connections. Simulating a range of states from wakefulness to sleep, the researchers found that this biologically grounded heterogeneity increased coordination between brain regions and improved information flow compared to a model in which all regions behaved identically.

"Whole-brain models offer systems neuroscientists deep insight into the global impact of local phenomena, giving us a better understanding of mechanisms and generating testable predictions. This study is an example of the impact of interareal heterogeneity on how global and local brain states are generated," says Maria V. Sanchez-Vives, researcher at IDIBAPS and last author of the study.

The model also spontaneously reproduced a phenomenon seen in real brains: localised, sleep-like slow waves appearing in some regions while the rest of the cortex stays in an awake-like state, previously observed during attentional lapses, sleep deprivation, and around brain lesions.

The findings shed light on how molecular-level detail can meaningfully shape brain-wide activity, offering a framework the authors suggest could eventually help explain state transitions in conditions such as brain lesions or disorders of consciousness.

The work also suggests that neuromodulators such as acetylcholine do not act uniformly across the brain. Instead, their effects depend on where their receptors are concentrated, meaning the same chemical signal can produce different network dynamics in different cortical regions – a finding that, combined with the underlying structural connectivity, could help future models capture how the brain shifts between wakefulness, sleep, and altered states.