Polygenic hierarchies of macroscale brain structural organisation
- 1University of Cambridge
Abstract
Different Magnetic Resonance Imaging-derived cortical, subcortical, and white matter measures are presumed to reflect different developmental and cellular processes, but how their genetic architecture is organised and the underlying cellular and developmental processes is unclear. We conducted genome-wide association studies of 2,326 imaging-derived phenotypes (IDPs) spanning twelve structural measures across the cortex, subcortex, ventricles, and white matter tracts in the UK Biobank (Nmax = 53,751), replicated in ABCD (Nmax = 5,119), identifying 14,176 experiment-wide significant loci and prioritizing 848 genes. We found that IDPs primarily cluster by measure into six families: brain size, cortical thickness, curvature, microstructural coherence, microstructural diffusivity, and orientation dispersion. This structure was preserved across prioritized genes, cell types, and developmental timing; brain size implicates first-trimester radial glia programs, while white matter coherence implicates adolescent astrocyte and oligodendrocyte lineages. Within individual measures, genetic effects were further organised along broad spatial gradients, with a principal gradient reflecting allometric scaling for most measures and secondary gradients showing correspondence with the sensory-association topographic axis across several measures. We identified 79 genes with spatially restricted effects along the cortex, with many aligning with these dominant gradients while a substantial subset showed additional spatial patterns. Together, these findings reveal a hierarchical organisation of the genetic architecture of macroscale human brain structure, spanning phenotype families, broad spatial gradients, and regionally restricted developmental programmes.
Related
- Distinct electroencephalography alpha-band rhythms map macroscale brain gradients
- Habitual sleep duration is associated with the cortical organisation of brain network controllability
- Development of Neonatal Functional Brain Network Organisation and Its Association with Childhood Neurodevelopmental Outcomes
- Attention Across Scales: From Individual Variation to Social Hierarchies and Brain Networks in Semi-Free-Ranging Macaques
- Non-Coding Structural Variants in Regulatory Regions and High Polygenic Burden May Jointly Contribute to Early-Onset Familial Hashimoto’s Thyroiditis
BibTeX
@misc{gu2026polygenic,
title = {{Polygenic hierarchies of macroscale brain structural organisation}},
author = {Y. Gu and A. Ebneabbasi and Y. He and C. Pecci-Terroba and K. Kumar and S. Jacquemonth and S. Baron-Cohen and H. Min and H. Won and E. Bullmore and R. Romero-Garcia and R. A. I. Bethlehem and V. Warrier},
howpublished = {bioRxiv},
year = {2026},
month = sep,
doi = {10.64898/2026.09.19.752830},
url = {https://www.biorxiv.org/content/10.64898/2026.09.19.752830v1},
note = {CC BY-NC-ND},
}
RIS
TY - UNPB
TI - Polygenic hierarchies of macroscale brain structural organisation
AU - Y. Gu
AU - A. Ebneabbasi
AU - Y. He
AU - C. Pecci-Terroba
AU - K. Kumar
AU - S. Jacquemonth
AU - S. Baron-Cohen
AU - H. Min
AU - H. Won
AU - E. Bullmore
AU - R. Romero-Garcia
AU - R. A. I. Bethlehem
AU - V. Warrier
PY - 2026
DA - 2026/09/19
PB - bioRxiv
DO - 10.64898/2026.09.19.752830
UR - https://www.biorxiv.org/content/10.64898/2026.09.19.752830v1
AB - Different Magnetic Resonance Imaging-derived cortical, subcortical, and white matter measures are presumed to reflect different developmental and cellular processes, but how their genetic architecture is organised and the underlying cellular and developmental processes is unclear. We conducted genome-wide association studies of 2,326 imaging-derived phenotypes (IDPs) spanning twelve structural measures across the cortex, subcortex, ventricles, and white matter tracts in the UK Biobank (Nmax = 53,751), replicated in ABCD (Nmax = 5,119), identifying 14,176 experiment-wide significant loci and prioritizing 848 genes. We found that IDPs primarily cluster by measure into six families: brain size, cortical thickness, curvature, microstructural coherence, microstructural diffusivity, and orientation dispersion. This structure was preserved across prioritized genes, cell types, and developmental timing; brain size implicates first-trimester radial glia programs, while white matter coherence implicates adolescent astrocyte and oligodendrocyte lineages. Within individual measures, genetic effects were further organised along broad spatial gradients, with a principal gradient reflecting allometric scaling for most measures and secondary gradients showing correspondence with the sensory-association topographic axis across several measures. We identified 79 genes with spatially restricted effects along the cortex, with many aligning with these dominant gradients while a substantial subset showed additional spatial patterns. Together, these findings reveal a hierarchical organisation of the genetic architecture of macroscale human brain structure, spanning phenotype families, broad spatial gradients, and regionally restricted developmental programmes.
ER -