Abstract
Several studies have suggested that functional connectivity (FC) is constrained by the underlying structural connectivity (SC) and mutually correlated. However, not many studies have focused on differences in the network organization of SC and FC, and on how these differences may inform us about their mutual interaction. To explore this issue, we adopt a multi-layer framework, with SC and FC, constructed using Magnetic Resonance Imaging (MRI) data from the Human Connectome Project, forming a two-layer multiplex network. In particular, we examine node strength assortativity within and between the SC and FC layer. We find that, in general, SC is organized assortatively, indicating brain regions are on average connected to other brain regions with similar node strengths. On the other hand, FC shows disassortative mixing. This discrepancy is apparent also among individual resting-state networks within SC and FC. In addition, these patterns show lateralization, with disassortative mixing within FC subnetworks mainly driven from the left hemisphere. We discuss our findings in the context of robustness to structural failure, and we suggest that discordant and lateralized patterns of associativity in SC and FC may provide clues to understand laterality of some neurological dysfunctions and recovery.
Highlights
The relationship between structural connectivity and functional connectivity has attracted much attention in recent years[1,2,3,4,5]
We first calculated node strength assortativity in each individual data set across the entire cerebral cortex
When we further investigated the above characteristics by separating the left and the right hemispheres, we found that the characteristic disassortative mixing in within functional connectivity (FC) layer is mainly driven by the left hemisphere (Figs 5 and 6)
Summary
The relationship between structural connectivity and functional connectivity has attracted much attention in recent years[1,2,3,4,5]. Not many studies have focused on fundamental topological differences in the network organization of SC and FC, and on how these differences may provide insight into their mutual interaction. To explore this issue, we adopt a multi-layer framework, with SC and FC forming a multiplex network[20,21,22]. If any systematic/consistent topological differences exist between these two networks, are they mere by-products of the generative process or could they represent biologically meaningful features of multi-layer organization that carry benefit or enhance overall functionality?. Correlations among nodes features can be measured both within- and between-layers[28,29]. Our findings may provide clues to understand why some brain networks are more vulnerable to or more resilient against functional disruption due to brain disorders or injuries
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call