The human brain is a complex network of structural and functional interconnections, with diverse regions activated during functional tasks. Advanced diffusion imaging methods, which track the diffuson of water along the brain’s axons, can reveal dense microstructural fiber bundles connecting anatomically distinct cortical and subcortical regions. Such connections are remodeled throughout development [1
] and deteriorate in diseases such as Alzheimer’s disease [2
]. While initial investigations have examined the degree of genetic involvement in functional connectivity, genetic contributions to the brain’s structural connectivity, i.e. the proportions and densities of axonal fibers connecting cortical subregions, have yet to be explored.
The degree of genetic influence on a particular trait can be determined by studying twins. Twin studies have long been used to determine the heritability (proportion of variance explainable by genetic variation) of human traits. Some studies have begun to estimate the heritability of DTI-derived measures of fiber integrity and its asymmetry as well as other neuroimaging measures [3
]. Proportions of variance due to genes versus environment can be inferred by fitting structural equation models (SEMs) to data from different types of twins--monozygotic (MZ) twins share all their genes while dizygotic (DZ) twins share, on average, half.
In a large family cohort comprised of 366 individuals from 223 families, we used high angular resolution diffusion imaging (HARDI) at high magnetic field (4 Tesla) along with anatomical MRI to delineate cortical regions into areas of known structure and functionality [7
]. We also mapped out white matter fiber pathways using high angular-resolution HARDI tractography. In this work, we define connectivity as the proportion of total fibers traced in the brain that intersect a specific pair of cortical regions – this may include connections within or between hemispheres. The connectivities of all pairs of regions are compiled into symmetric matrices, in which each matrix element (x,y
) is the proportion of fibers connecting brain regions x
To determine the genetic influences contributing to the density of each cortical connection, we fitted a SEM to connectivity matrices extracted from 46 pairs of MZ and 64 pairs of DZ twins. If a connection was significantly influenced by genetic factors, we followed through with a genome-wide association test to identify specific genetic variants associated with the proportion of fiber densities at that connection. The sample was split in half to allow replication of discovered associations in non-overlapping samples.