A Large-Scale Genetic Correlation Scan Between Intelligence and Brain Imaging Phenotypes

• Published in: Cerebral cortex (New York, N.Y. 1991) Vol. 30; no. 7; pp. 4197 - 4203
• Main Authors: Cheng, Shiqiang, Wu, Cuiyan, Qi, Xin, Liu, Li, Ma, Mei, Zhang, Lu, Cheng, Bolun, Liang, Chujun, Li, Ping, Kafle, Om Prakash, Wen, Yan, Zhang, Feng
• Format: Journal Article
• Language: English
• Published: United States Oxford University Press 01.06.2020
• Subjects: Adult; Aged; Brain - diagnostic imaging; Brain - physiology; Female; Functional Neuroimaging; Genome-Wide Association Study; Humans; Intelligence - genetics; Magnetic Resonance Imaging; Male; Middle Aged; Phenotype; polygenic risk score; linkage disequilibrium score regression; brain image-derived phenotypes; intelligence
• ISSN: 1047-3211; 1460-2199; 1460-2199
• DOI: 10.1093/cercor/bhaa043

Abstract Limited efforts have been paid to evaluate the potential relationships between structural and functional brain imaging and intelligence until now. We performed a two-stage analysis to systematically explore the relationships between 3144 brain image-derived phenotypes (IDPs) and intelligence. First, by integrating genome-wide association studies (GWAS) summaries data of brain IDPs and two GWAS summary datasets of intelligence, we systematically scanned the relationship between each of the 3144 brain IDPs and intelligence through linkage disequilibrium score regression (LDSC) analysis. Second, using the individual-level genotype and intelligence data of 160 124 subjects derived from UK Biobank datasets, polygenetic risk scoring (PRS) analysis was performed to replicate the common significant associations of the first stage. In the first stage, LDSC identified 6 and 2 significant brain IDPs significantly associated with intelligence dataset1 and dataset2, respectively. It is interesting that NET100_0624 showed genetic correlations with intelligence in the two datasets of intelligence. After adjusted for age and sex as the covariates, NET100_0624 (P = 5.26 × 10−20, Pearson correlation coefficients = −0.02) appeared to be associated with intelligence by PRS analysis of UK Biobank samples. Our findings may help to understand the genetic mechanisms of the effects of brain structure and function on the development of intelligence.