Genome-wide mediation analysis of psychiatric and cognitive traits through imaging phenotypes

Abstract

Heritability is well documented for psychiatric disorders and cognitive abilities which are, however, complex, involving both genetic and environmental factors. Hence, it remains challenging to discover which and how genetic variations contribute to such complex traits. In this article, they propose to use mediation analysis to bridge this gap, where neuroimaging phenotypes were utilized as intermediate variables. The Philadelphia Neurodevelopmental Cohort was investigated using genome-wide association studies (GWAS) and mediation analyses. Specifically, 951 participants were included with age ranging from 8 to 21 years. Two hundred and four neuroimaging measures were extracted from structural magnetic resonance imaging scans. GWAS were conducted for each measure to evaluate the SNP-based heritability. Furthermore, mediation analyses were employed to understand the mechanisms in which genetic variants have influence on pathological behaviors implicitly through neuroimaging phenotypes, and identified SNPs that would not be detected otherwise. Our analyses found that rs10494561, located in the intron region within NMNAT2, was associated with the severity of the prodromal symptoms of psychosis implicitly, mediated through the volume of the left hemisphere of the superior frontal region ( P=2.38×10-8). The gene NMNAT2 is known to be associated with brainstem degeneration, and produce cytoplasmic enzyme which is mainly expressed in the brain. Another SNP rs2285351 was found in the intron region of gene IFT122 which may be potentially associated with human spatial orientation ability through the area of the left hemisphere of the isthmuscingulate region ( P=3.70×10-8). Hum Brain Mapp 38:4088-4097, 2017. © 2017 Wiley Periodicals, Inc.

Keywords: genome-wide association studies; imaging genetics; intermediate phenotypes; neuroimaging measures.

Figure 1

Mediation analysis. A mediation model was utilized to detect the direct and indirect effect that a genetic variant may have on a cognitive or psychiatric trait. a): Genome‐wide association scans were performed to search for pairs of intermediate neuroimaging phenotypes and genetic variants that had significant associations. b): The cognitive or psychiatric trait was fit against each candidate genetic variant to test for direct and significant influence. c): The cognitive or psychiatric trait was fit against identified genetic variant and its associated intermediate neuroimaging phenotype simultaneously. A mediation relationship is built if 1): the genetic variant is significant in a), 2): the genetic variant is significant in b), 3): the intermediate neuroimaging phenotype is significant in c), while the genetic variant has a smaller coefficient magnitude in c) than in a). A mediation relationship indicates a mechanism that the genetic variant may have implicit influence on the trait through the intermediate neuroimaging phenotype. [Color figure can be viewed at http://wileyonlinelibrary.com]

Figure 2

GWAS were conducted through linearly regressing the superior frontal volume and the isthmus cingulate area of the left hemisphere against each SNP, controlling for age, sex and the first two principal components of the genotype data. QQ plots for the two GWAS were illustrated on the left and right panels. The observed P‐values were plotted against the expected P‐values, after a transformation of negative ten‐based logarithm. The red straight lines had slope equal to 1. The genomic inflation factors were 1.013672 and 1.011903 for the left and the right panels, respectively. [Color figure can be viewed at http://wileyonlinelibrary.com]

Figure 3

GWAS were conducted through linearly regressing the superior frontal volume and the isthmus cingulate area of the left hemisphere of the left hemisphere against each SNP, controlling for age, sex, and the first two principal components of the genotype data. Manhattan plot for each of the two measures were illustrated in the top and the bottom panel, where the blue horizontal line represented the multiple‐comparisons‐corrected threshold of $P=1.12\times {10}^{-7}$, and the red horizontal line represented the common threshold of $P=5\times {10}^{-8}$.