Detection of circular RNA expression and related quantitative trait loci in the human dorsolateral prefrontal cortex

BACKGROUND: Circular RNAs (circRNAs) are implicated in various biological processes. As a layer of the gene regulatory network, circRNA expression is also an intermediate phenotype bridging genetic variation and phenotypic changes. Thus, analyzing circRNA expression variation will shed light on molecular fundamentals of complex traits and diseases. RESULTS: We systematically characterize 10,559 high-quality circRNAs in 589 human dorsolateral prefrontal cortex samples. We identify biological and technical factors contributing to expression heterogeneity associated with the expression levels of many circRNAs, including the well-known circRNA CDR1as. Combining the expression levels of circRNAs with genetic cis-acting SNPs, we detect 196,255 circRNA quantitative trait loci (circQTLs). By characterizing circQTL SNPs, we find that partial circQTL SNPs might influence circRNA formation by altering the canonical splicing site or the reverse complementary sequence match. Additionally, we find that a subset of these circQTL SNPs is highly linked to genome-wide association study signals of complex diseases, especially schizophrenia, inflammatory bowel disease, and type II diabetes mellitus. CONCLUSIONS: Our results reveal technical, biological, and genetic factors affecting circRNA expression variation among individuals, which lead to further understanding of circRNA regulation and thus of the genetic architecture of complex traits or diseases. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (10.1186/s13059-019-1701-8) contains supplementary material, which is available to authorized users.