Preprint: non-SNP variants account for over half of regulatory genetic associations in rat brain eQTL study

Researchers have posted a preprint on bioRxiv reporting a large-scale genetic and transcriptomic study in Heterogeneous Stock (HS) rats, an outbred population widely used as an animal model for complex traits including addiction-related behaviours. The study used both short-read and long-read sequencing to generate a comprehensive catalogue of genetic variants in this population, encompassing single nucleotide polymorphisms (SNPs), small insertions and deletions (indels), short tandem repeats, and structural variants.

The team then performed joint cis-expression quantitative trait locus (cis-eQTL) mapping across five brain regions, asking which variants most strongly associate with local gene expression differences. A notable finding reported in the lede is that non-SNP variants — including structural variants and repeat-length polymorphisms — accounted for over 50% of lead regulatory associations, and that many of these were poorly tagged by SNPs alone, suggesting that standard SNP-array-based analyses may underestimate the contribution of these variant classes to gene regulation in the brain.

The work has implications for understanding the genetic architecture of complex brain traits and for the design and interpretation of genetic studies of addiction and related behavioural phenotypes. It also illustrates the value of long-read sequencing in capturing variant classes missed by short-read approaches. The preprint is not yet peer-reviewed. It will be of primary interest to statistical geneticists, researchers in neuroscience genomics, and those working on eQTL methodology.