Aguirre Matthew, Spence Jeffrey P, Sella Guy, Pritchard Jonathan K
Department of Biomedical Data Science, Stanford University, Stanford CA.
Department of Genetics, Stanford University, Stanford CA.
bioRxiv. 2025 Aug 12:2025.08.12.669924. doi: 10.1101/2025.08.12.669924.
In human populations, most of the genetic variance in gene expression can be attributed to -acting expression quantitative trait loci (eQTLs) spread across the genome. However, in practice it is difficult to discover these eQTLs, and their cumulative effects on gene expression and complex traits are yet to be fully understood. Here, we assess how properties of the genetic architecture of gene expression constrain the space of plausible gene regulatory networks. We describe a structured causal model of gene expression regulation and consider how it interacts with biologically relevant properties of the gene regulatory network to alter the genomic distribution of expression heritability. Under our model, we find that the genetic architecture of gene expression is shaped in large part by local network motifs and by hub regulators that shorten paths through the network and act as key sources of -acting variance. Further, simulated networks with an enrichment of motifs and hub regulators best recapitulate the distribution of and heritability of gene expression as measured in a recent twin study. Taken together, our results suggest that the architecture of gene expression is sparser and more pleiotropic across genes than would be suggested by naive models of regulatory networks, which has important implications for future studies of complex traits.
在人类群体中，基因表达的大部分遗传变异可归因于分布在基因组中的顺式作用表达数量性状位点（eQTL）。然而，在实际中很难发现这些eQTL，而且它们对基因表达和复杂性状的累积影响尚未得到充分理解。在这里，我们评估基因表达的遗传结构特性如何限制合理的基因调控网络空间。我们描述了一个基因表达调控的结构化因果模型，并考虑它如何与基因调控网络的生物学相关特性相互作用，以改变表达遗传力的基因组分布。在我们的模型下，我们发现基因表达的遗传结构在很大程度上由局部网络基序和枢纽调节因子塑造，这些枢纽调节因子缩短了通过网络的路径，并作为顺式作用变异的关键来源。此外，具有丰富基序和枢纽调节因子的模拟网络最能重现最近一项双胞胎研究中测得的基因表达的加性和显性遗传力分布。综上所述，我们的结果表明，基因表达的结构在基因间比简单的调控网络模型所暗示的更为稀疏和多效，这对未来复杂性状的研究具有重要意义。
