Principal and independent genomic components of brain structure and function

It is commonly accepted that our genes influence the shape of our nose and the color of our eyes, and our genes also play an important role in the shape and functioning of our brain. However, the complex biological mechanisms that cause individual differences in the brain remain largely unknown. Genome-wide association studies (GWAS) investigate common genetic variants along the entire length of the DNA in association with specific traits or diseases. GWASs have revealed that thousands of genetic variants are associated with individual differences in brain structure and function, behaviors and brain-related conditions. However, GWASs results represent the endpoints of many causal biological processes. The processes by which DNA variation influences brain structure and function include gene expression, protein synthesis, cellular functioning, brain morphology and functioning, and environmental factors. The involvement of many genetic variants in the development of a trait, paired with the abundance of biological processes at work, make it difficult to infer the causal processes that drive the genetic associations. Therefore, deriving a mechanistic understanding of the inheritance of brain-related variation from GWAS output, is very difficult.

Researchers Lennart Oblong and Sourena Soheili-Nezhad have worked to improve upon the interpretability of genome-wide associations, by developing a new statistical method that analyzes the output of many brain-related GWASs. Led by Emma Sprooten, the research group for Psychiatry and Neuroimaging Genetics published their new method in the journal Genes, Brain and Behavior on 15 January.

In their paper, a statistical method was developed and optimized to process the genome-wide association data of over 2000 different brain variables simultaneously. Included variables were for example gray matter volumes of various regions of the brain, white matter properties connecting these regions, and functional brain activation patterns. This new method, called genomic independent component analysis (genomic ICA), can transform the vast gene-brain association data into fewer, more interpretable components (i.e. factors). These new components reflect the similarity of genetic associations across many brain phenotypes, and thereby tell us more about which genes might work together to influence specific brain phenotypes. The reproducibility of the resultant independent and principal components was tested across independent samples. The outcome was promising, as these components exceeded reproducibility metrics achieved previously, and showed clear patterns of genetic variants associated with distinct brain features.

It is reasonable to assume that the brain plays a crucial role in shaping our mental faculties and moods, with neuroimaging research highlighting the connection between brain variations and mental health traits. However, the manifold influence of genetic variants on the brain, along with the etiological heterogeneity, suggesting that factors vary significantly among individuals in different contexts, should not be underestimated in mental health related studies. To tackle this complexity, the team is presently working on using the new genomic ICA method, and its resultant genomic components, to gain new insights into which genetic variations and biological processes are significant for individuals in different environmental contexts, with the goal of improving their mental health. Ultimately, in collaboration with the FAMILY consortium, our goal is to apply this method to enhance our understanding of how the shared effects of many genetic variants influence the development of brain structure and mental health conditions within families. This approach may thus be a stepping stone to gain a better understanding of resilience to mental illness, against the background of genetic risk factors.