A vulnerability analysis of critical brain regions and connections in the mouse connectome

The human brain contains billions of neurons with trillions of complex connections among them. To understand this complexity, the field of connectomics is mapping and analyzing the nervous systems of many different animal species to help elucidate the fundamental principles of brain function. Focusing on the mammalian brain, in this study we analyzed the mouse connectome using a graph theoretical approach to uncover the critical brain regions and connections of the network. Vulnerability analysis conducts simulated deletions of each region and connection to measure each component's specific contributions to the brain as an entire system - thus it not only captures the local network properties of each component, but their effects on other regions and connections as well. Our analysis identified 8 regions and 38 connections as being critical to the network information processing. We also found that most of the critical connections and brain regions were interconnected with each other. Moreover, this interconnectivity organized into four main groups that can be categorized based on known neurobiological function: most notably, olfaction, learning and memory, higher-level motor control and cognition, and lower-level autonomic and motor control. The vulnerability result thus suggests that these functional components appear to be the most fundamental and thus most important for the mouse's survival. The findings of this study suggest that the vulnerability analysis approach can be used to uncover and characterize brain complexity, including ultimately the human brain.