Using transmitted and untransmitted gene networks to identify molecular pathways to substance use & misuse in genetically controlled twins

点击翻译按钮获取中文摘要

Project Summary Multiple genetic risks cause alcohol, nicotine and cannabis substance use (SU) and substance use disorders (SUDs). However, genetic studies alone cannot explain the cascade of molecular changes between SNPs and SU and SUDs. Nor can current approaches correctly measure the causal impact of genetic differences on parental and home environments linked to SU and SUDs. Molecular studies that link GWAS results to gene expression data via eQTL findings have increased our understanding of the etiology of SU and SUDs. Yet, they fail to capture complex gene interactions that are the hallmark of complex traits including SU and SUDs, or are mostly underpowered (especially from brain tissues). Another limitation is that genetic and molecular studies neglect environmental confounding. We know that parents create and influence children’s environments and that certain parental and home environments are more associated with SU and SUDs. Not only do genetic risks vary with environments, but environments themselves are also heritable, and are thus confounded by parental genetics. We and others argue that unconfounding the impact of parent-to-offspring molecular genetics on SU and SUDs from the parental and home environments is the central task of genetics. Here, we rely on recent methodological advances to address these limitations. First, we have developed a novel and proven method of measuring the impact of untransmitted parental genomes. This method identifies the indirect impact of untransmitted parental genomes on complex behaviours. A significant impact of untransmitted genomes is evidence of genetically nurtured environments that are unconfounded by the direct parent-to-offspring genetics. Second, we can now perform gene expression (GE) imputation using existing GWAS. We can, therefore, estimate heritable, genome-wide individual differences in imputed GE networks in very large samples. This substantially increases our power to identify molecular pathways underlying SU and SUD aetiology. In summary, we are proposing to i) use existing GWAS data from large population-based family samples, ii) apply our method of assembling untransmitted parental genomes, iii) impute parentally transmitted and untransmitted GE, iv) apply network analyses to identify transmitted and untransmitted GE networks that are associated with SU and SUDs, and v) identify parental and home environments mediating these networks. This will enable us to identify the GE networks involved in SU and SUDs while identifying risky and protective parental and home environments that are genetically ‘nurtured’ within a unified theoretical framework.