Genome-wide association study of the human brain functional connectome reveals strong vascular component underlying global network efficiency

Abstract

Complex brain networks play a central role in integrating activity across the human brain, and such networks can be identified in the absence of any external stimulus. We performed 10 genome-wide association studies of resting state network measures of intrinsic brain activity in up to 36,150 participants of European ancestry in the UK Biobank. We found that the heritability of global network efficiency was largely explained by blood oxygen level-dependent (BOLD) resting state fluctuation amplitudes (RSFA), which are thought to reflect the vascular component of the BOLD signal. RSFA itself had a significant genetic component and we identified 24 genomic loci associated with RSFA, 157 genes whose predicted expression correlated with it, and 3 proteins in the dorsolateral prefrontal cortex and 4 in plasma. We observed correlations with cardiovascular traits, and single-cell RNA specificity analyses revealed enrichment of vascular related cells. Our analyses also revealed a potential role of lipid transport, store-operated calcium channel activity, and inositol 1,4,5-trisphosphate binding in resting-state BOLD fluctuations. We conclude that that the heritability of global network efficiency is largely explained by the vascular component of the BOLD response as ascertained by RSFA, which itself has a significant genetic component.

Disentangling associations between complex traits and cell types withseismic

“…Comparing cell types that overlap between TM FACS and these 2 additional scRNA-seq datasets, one using a different technology, and the other using human cells, we find consistent cell type-trait associations (Supplementary Figures 9,11,12, mean Spearman's correlation of 0.79 between TM FACS and TM droplet, and 0.67 between TM FACS and TS; we note that neither TM droplet nor TS contain brain tissue, and the mean Spearman's correlation is 0.84 and 0.75 respectively if neuropsychiatric traits are excluded from the comparison). The high concordance of seismic with other methods is also consistent across datasets (Supplementary Figures 10,13,14).…”

“…9,11,12, mean Spearman's correlation of 0.79 between TM FACS and TM droplet, and 0.67 between TM FACS and TS; we note that neither TM droplet nor TS contain brain tissue, and…”

“…Through a deep exploration of neurological disease-associated brain cell types, we find that cell type definition from input scRNA-seq data is an important, underappreciated, factor that influences downstream findings. Specifically, previous analyses [6, 8, 11, 12] have typically used broad characterizations of cell types such as “telencephalon projecting excitatory neurons” and “frontal cortex neurons,” which do not consider additional region or tissue granularity, possibly obscuring valuable biological insights. For example, in Alzheimer’s disease, a population of neurons in the entorhinal cortex are most vulnerable, while neurons in even neighboring brain regions such as the dentate gyrus and CA2/CA3 in the hippocampus are not [13].…”

Uncovering the multivariate genetic architecture of frailty with genomic structural equation modelling

“…Furthermore, since this represented the first time that frailty has been measured in this latent framework, we wanted to validate our factors as reflecting frailty since well-powered independent GWAS samples for all 30 deficits are not currently available to do a full replication GWAS. Therefore, we performed genetic correlation analyses to assess the associations between 52 aging-related health outcomes and frailty phenotypes to assess the different patterns of shared genetics between these outcomes and each of the latent factors (Table S4-S5) (20,21,(71)(72)(73)(74)(75)(76)(77)(78)(79)(80)(81)(82)(83)(84)(85)(86)(87)(88)(89)(90). We used the same quality control procedures and data curation steps on the GWAS summary statistics for these 52 health outcomes as described for the main frailty deficits using the munge function in the GenomicSEM R package.…”