A geographic cline of skull and brain morphology among individuals of European Ancestry

Abstract

Background: Human skull and brain morphology are strongly influenced by genetic factors, and skull size and shape vary worldwide. However, the relationship between specific brain morphology and genetically-determined ancestry is largely unknown.

Methods: We used two independent data sets to characterize variation in skull and brain morphology among individuals of European ancestry. The first data set is a historical sample of 1,170 male skulls with 37 shape measurements drawn from 27 European populations. The second data set includes 626 North American individuals of European ancestry participating in the Alzheimer's Disease Neuroimaging Initiative (ADNI) with magnetic resonance imaging, height and weight, neurological diagnosis, and genome-wide single nucleotide polymorphism (SNP) data.

Results: We found that both skull and brain morphological variation exhibit a population-genetic fingerprint among individuals of European ancestry. This fingerprint shows a Northwest to Southeast gradient, is independent of body size, and involves frontotemporal cortical regions.

Conclusion: Our findings are consistent with prior evidence for gene flow in Europe due to historical population movements and indicate that genetic background should be considered in studies seeking to identify genes involved in human cortical development and neuropsychiatric disease.

Fig. 1

Cranial morphology reflects geography across Europe. a Pairwise distances between 27 European populations. Craniometric distance is significantly correlated (r_M = 0.51, p < 1 × 10^-5) with geographic distance. b Non-metric multi-dimensional scaling ordination of craniometric distances aligned to geographic coordinates of populations. Population symbols identify 4 clusters, and lines form a minimum spanning tree. c Distances between predicted locations based on craniometric ordination and population locations. Average ± SD plotted for 100 bootstrap replications (black) and random permutations (gray). * p < 0.001. d Individual female skulls were identified with correct or nearby populations based on cranial morphometry (solid) significantly better than chance (dotted). e Proportion of female skulls that were correctly classified (black) and misclassified with populations at different distances (gray shades). Sample sizes are listed after population name.

Fig. 2

Cranial measures show significant variation along a NWSE axis within Europe. a Isoclines of interpolated eigenvalues for first spatially constrained component of a redundancy analysis and geographic locations of populations. b Cranial measures plotted in order of their contribution to this map. Negative abscissas correspond to a more NW location. Proportion of variance explained (R²) and nominal p values are indicated. NLB = Nasal breadth; M28 = sagittal occipital arc; GOL = glabello-occipital length; NOL = nasio-occipital length; ASB = biasterionic breadth; BBH = basion-bregma height.

Fig. 3

Structural brain measures follow a predicted trend in a group of individuals with European ancestry. a The first two principal components of genotypes of ADNI subjects (yellow/small points; color refers to online version only) and individuals from European reference populations (gray crosses) rotated 18° to align with a map of Europe. For each reference population (see online suppl. table S3 for labels), the average (SD) of principal components for all individuals in that population are indicated by disc position (diameter). Geographic origin of each population is indicated by disc shade of gray from NW (black) to SE (light gray) Europe. ADNI subjects are spread out primarily along a NW-SE axis and form two distinct clusters corresponding to NW European and Ashkenazi Jewish ancestry (see also online suppl. fig. S5). b Brain structural measures tested for association with estimated NW-SE ancestry, while controlling for height, weight, BMI, age, sex, and diagnosis. Negative abscissas correspond to a larger proportion of NW ancestry.

Fig. 4

Frontotemporal cortical regions are most affected by NW European ancestry. Lateral view of the left hemisphere with color map that indicates nominal -log₁₀ (p value) of association between estimated NW-SE ancestry and cortical surface area across the reconstructed cortical surface, while controlling for height, weight, BMI, age, sex, and diagnosis.