|
Abstract:
|
It is of great interest to quantify the heritability of brain structure and function. Decomposing brain traits into genetic and environmental effects can inform our understanding of human behavior and mental health. We develop a method to estimate heritability and the non-stationary covariance components in high-dimensional imaging data from twin studies. Our motivating example is from the Human Connectome Project (HCP). We extend Fisher’s additive genetic, common environmental, and unique environmental model to exploit spatial smoothness across brain surfaces, which we term the ACE of Space. We apply this model to estimate heritability of cortical thickness, which reveals fine-scale spatial differences in genetic control. We also propose a method for functional connectivity correlation matrices. Our model and algorithm ensure positive semidefiniteness of the resulting covariance matrices of genetic effects, which can improve estimates of heritability.
|
