关键词: Mendelian randomization biomarker brain imaging–derived phenotypes cognitive functions verbal–numerical reasoning

Mesh : Humans Cognition / physiology Phenotype Brain / diagnostic imaging physiology Genome-Wide Association Study Mendelian Randomization Analysis Male Female Neuroimaging / methods Middle Aged Magnetic Resonance Imaging / methods Aged

来  源:   DOI:10.1093/cercor/bhae297

Abstract:
We aimed to evaluate the potential causal relationship between brain imaging-derived phenotypes and cognitive functions via Mendelian randomization analyses. Genetic instruments for 470 brain imaging-derived phenotypes were selected from a genome-wide association study based on the UK Biobank (n = 33,224). Statistics for cognitive functions were obtained from the genome-wide association study based on the UK Biobank. We used the inverse variance weighted Mendelian randomization method to investigate the associations between brain imaging-derived phenotypes and cognitive functions, and reverse Mendelian randomization analyses were performed for significant brain imaging-derived phenotypes to examine the reverse causation for the identified associations. We identified three brain imaging-derived phenotypes to be associated with verbal-numerical reasoning, including cortical surface area of the left fusiform gyrus (beta, 0.18 [95% confidence interval, 0.11 to 0.25], P = 4.74 × 10-7), cortical surface area of the right superior temporal gyrus (beta, 0.25 [95% confidence interval, 0.15 to 0.35], P = 6.30 × 10-7), and orientation dispersion in the left superior longitudinal fasciculus (beta, 0.14 [95% confidence interval, 0.09 to 0.20], P = 8.37 × 10-7). The reverse Mendelian randomization analysis indicated that verbal-numerical reasoning had no effect on these three brain imaging-derived phenotypes. This Mendelian randomization study identified cortical surface area of the left fusiform gyrus, cortical surface area of the right superior temporal gyrus, and orientation dispersion in the left superior longitudinal fasciculus as predictors of verbal-numerical reasoning.
摘要:
我们旨在通过孟德尔随机化分析来评估脑成像表型与认知功能之间的潜在因果关系。从基于UKBiobank的全基因组关联研究中选择了470种脑成像衍生表型的遗传仪器(n=33,224)。认知功能的统计数据来自基于UKBiobank的全基因组关联研究。我们使用逆方差加权孟德尔随机化方法来研究脑成像衍生的表型与认知功能之间的关联。并对重要的脑成像衍生表型进行了反向孟德尔随机化分析,以检查已确定关联的反向因果关系.我们确定了三种脑成像衍生的表型与言语数字推理有关,包括左梭状回的皮质表面积(β,0.18[95%置信区间,0.11to0.25],P=4.74×10-7),右颞上回的皮质表面积(β,0.25[95%置信区间,0.15至0.35],P=6.30×10-7),和左上纵束中的方向分散(β,0.14[95%置信区间,0.09to0.20],P=8.37×10-7)。反向孟德尔随机化分析表明,言语数字推理对这三种脑成像衍生的表型没有影响。这项孟德尔随机研究确定了左梭状回的皮质表面积,右颞上回的皮质表面区域,和左上纵束中的方向分散作为言语数字推理的预测因子。
公众号