Gene function & annotation

Structural variants are associated with changes in gene expression in blood of Holstein cows

---

Abstract

Structural variants (SVs) are large genomic rearrangements (≥50bp), including deletions, insertions, duplications, and inversions. Long-read sequencing has greatly improved the discovery and genotyping of SVs in cattle genomes. However, their functional impact on gene expression remains poorly understood. Although we can discover genomic positions of SVs using long read sequences and predict their impact within genomic regions with Variant Effect Predictor (VEP), their effect on gene expression is only known for a few high-profile SVs. Also, software such as VEP provides in silico prediction of SVs based on position and is not good at predicting the impact of intergenic variants. In this study, we aimed to investigate the effects of SVs on nearby gene expression by integrating long-read-based SV genotypes with RNA -sequence (RNA-seq) data from a small cattle population. Oxford Nanopore Technologies (ONT) long-read sequencing data of 160 cattle (86 Holstein and 74 Jersey) was used to generate a comprehensive catalogue of genome-wide SVs with Sniffles2 software in join-calling mode. Of these 160 cattle there were 28 Holsteins that also had RNA-seq data. Therefore, the segregating, biallelic SVs in 28 Holstein cattle with their genotypes were extracted, including 16,660 deletions, 17,851 insertions and 117 duplications. Gene expression levels were quantified from RNA-seq data from whole blood sampled from the same animals in early lactation (< 14 days in milk). The original read counts obtained from FeatureCounts were transformed into normalized logCPM values and adjusted for differences in library size among samples. For each SV and overlapping or nearby genes (within ±1 Mb), Wald tests were used to identify potential associations between SV genotypes and gene expression counts. Among the total of 450,125 SV-gene pairs tested, 1,398 SV-gene pairs showed association after multiple testing correction (FDR-adjusted p-value < 0.05). Fifteen SVs were in coding regions and were annotated as "high" impact by VEP. Sixty-one of the associations involved SVs in intronic regions, while others were located upstream or downstream of genes. We identified a list of top SV-gene pairs with potential regulatory functions and compared them with previously reported quantitative trait loci (QTLs) in cattle. Some identified SVs co-localized with known QTLs; for example, a duplication on BTA18 containing the whole CA5A gene overlapped with QTLs associated with blood calcium regulation in Holstein cattle, and animals carrying this duplication showed a significant increase in CA5A expression in our analysis, suggesting it may be a potential candidate for future functional validation. A larger number of animals with RNA-seq from the same experiment have been imputed for these SVs, and this data will be used in in the next steps to further confirm these results.

Keywords: 2026

How to Cite:

Tran, H., Macleod, I., Nguyen, T., Wang, J., Tahir, M., Thakur, P., Reich, C. & Chamberlain, A., (2026) “Structural variants are associated with changes in gene expression in blood of Holstein cows”, World Congress on Genetics Applied to Livestock Production Digital Archive 2026(1): 2286259. doi: https://doi.org/10.31274/wcgalp.23898

Rights: 1