Luciano Michelle, Hagenaars Saskia P, Cox Simon R, Hill William David, Davies Gail, Harris Sarah E, Deary Ian J, Evans David M, Martin Nicholas G, Wright Margaret J, Bates Timothy C
Department of Psychology, School of Philosophy, Psychology and Language Sciences, Centre for Cognitive Ageing and Cognitive Epidemiology, The University of Edinburgh, 7 George Square, Edinburgh, EH8 9JZ, UK.
Division of Psychiatry, University of Edinburgh, Edinburgh, EH10 5HF, UK.
Behav Genet. 2017 Sep;47(5):469-479. doi: 10.1007/s10519-017-9859-x. Epub 2017 Jul 15.
Impairments in reading and in language have negative consequences on life outcomes, but it is not known to what extent genetic effects influence this association. We constructed polygenic scores for difficulties with language and learning to read from genome-wide data in ~6,600 children, adolescents and young adults, and tested their association with health, socioeconomic outcomes and brain structure measures collected in adults (maximal N = 111,749). Polygenic risk of reading difficulties was associated with reduced income, educational attainment, self-rated health and verbal-numerical reasoning (p < 0.00055). Polygenic risk of language difficulties predicted income (p = 0.0005). The small effect sizes ranged 0.01-0.03 of a standard deviation, but these will increase as genetic studies for reading ability get larger. Polygenic scores for childhood cognitive ability and educational attainment were correlated with polygenic scores of reading and language (up to 0.09 and 0.05, respectively). But when they were included in the prediction models, the observed associations between polygenic reading and adult outcomes mostly remained. This suggests that the pathway from reading ability to social outcomes is not only via associated polygenic loads for general cognitive function and educational attainment. The presence of non-overlapping genetic effect is indicated by the genetic correlations of around 0.40 (childhood intelligence) and 0.70 (educational attainment) with reading ability. Mendelian randomization approaches will be important to dissociate any causal and moderating effects of reading and related traits on social outcomes.
阅读和语言障碍会对生活结果产生负面影响,但尚不清楚基因效应在多大程度上影响这种关联。我们利用约6600名儿童、青少年和年轻人的全基因组数据构建了语言和阅读学习困难的多基因分数,并测试了它们与成年人收集的健康、社会经济结果和脑结构测量指标之间的关联(最大样本量N = 111749)。阅读困难的多基因风险与收入降低、教育程度、自评健康状况以及言语数字推理能力相关(p < 0.00055)。语言困难的多基因风险可预测收入(p = 0.0005)。效应量较小,范围为标准差的0.01 - 0.03,但随着阅读能力基因研究样本量的增大,这些效应量将会增加。儿童认知能力和教育程度的多基因分数与阅读和语言的多基因分数相关(分别高达0.09和0.05)。但当将它们纳入预测模型时,多基因阅读与成人结果之间观察到的关联大多仍然存在。这表明从阅读能力到社会结果的途径不仅是通过一般认知功能和教育程度的相关多基因负荷。阅读能力与儿童期智力(约0.40)和教育程度(约0.70)的遗传相关性表明存在非重叠的基因效应。孟德尔随机化方法对于区分阅读及相关性状对社会结果的任何因果和调节效应将至关重要。