Abstract:The human connectome is the set of physical pathways linking brain regions to one another. Empirical and in silico studies have demonstrated that the structure of this network helps shape patterns of functional coupling between brain regions. To better understand this link between structure and function, a growing number of studies have derived geometric, dynamic, and topological predictors from structural connectivity in order to make predictions about correlation structure. These studies, however, have typically focused on global (whole-brain) predictions using a restricted set of predictors. Here, we investigate a wide range of predictors and shift focus onto predictions of local (regional) patterns of functional coupling. We show that, globally, no individual predictor performs well and, that even the best predictors are largely driven by their ability to predict functional coupling between directly connected regions. We then use the same predictors to make predictions of local coupling and find marked improvement. Notably, the most predictable local FC is linked to sensorimotor regions, which are best predicted by measures based on topological similarity, mean first passage times of random walkers, and the brain’s embedding in Euclidean space. We then show that by combining the predictive power of more than one predictor using multi-linear models, we can further improve local predictions. Finally, we investigate how global and local structure-function coupling differences across the human lifespan. We find that, globally, the magnitude of coupling decreases with biological age, which is paralleled by an increase in the number of multi-step pathways. We also show that, locally, structure-function coupling is preserved in higher order cognitive systems, but preferentially decreases with age in sensorimotor systems. Our results illuminate the heterogeneous landscape of structure-function coupling across the cerebral cortex and help clarify its differences with age.

Mapping the structure-function relationship along macroscale gradients in the human brain

MULTIMODAL GRADIENTS UNIFY LOCAL AND GLOBAL CORTICAL ORGANIZATION

Functional Organization of the Human Corpus Callosum Unveiled with BOLD-fMRI Gradients

Brain structure-function relationships across the human lifespan based on network eigenmodes

Decoupling of brain function from structure reveals regional behavioral specialization in humans

Myelination and excitation-inhibition balance synergistically shape structure-function coupling across the human cortex

Mapping the Structural Core of Human Cerebral Cortex

Genetic and phylogenetic uncoupling of structure and function in human transmodal cortex

Network structure of cerebral cortex shapes functional connectivity on multiple time scales

Cortical Gradient of a Human Functional Similarity Network Captured by the Geometry of Cytoarchitectonic Organization

A multi-scale cortical wiring space links cellular architecture and functional dynamics in the human brain

Structural mediation of human brain activity revealed by white-matter interpolation of fMRI

Combining function and structure in a single macro-scale connectivity model of the human brain

Local structure-function relationships in human brain networks across the lifespan

Brain topography beyond parcellations: Local gradients of functional maps

Mapping the Microstructure of Human Cerebral Cortex In Vivo with Diffusion MRI

Brain Cortical Functional Gradients Predict Cortical Folding Patterns via Attention Mesh Convolution

Brain connectomes come of age

Modeling brain signaling as Markovian helps explain its structure-function relationship

Structural Brain Architectures Match Intrinsic Functional Networks and Vary Across Domains: A Study from 15 000+ Individuals.

The morphospace of the brain-cognition organisation