Assessing the diversity and functional consequences of multiallelic VNTRs in cattle with long-read sequencing
Abstract
Variable number tandem repeats (VNTRs) are tandemly repeated sequences that vary in number of repeat motifs and/or in the sequence of the repeat motif. These neighboring, near identical, repeat sequences make sequence alignments difficult. This is particularly true with short read sequencing where individual reads cannot span larger tandemly repeated sequences. Long read sequencing technologies now provide individual reads that span many of these loci including their unique flanking sequences. These data and specialized genotyping programs like the Tandem Repeat Genotyping Tool (TRGT) facilitate genotyping of VNTRs. TRGT requires a repeat definition file as input which annotates tandem repeat (TR) loci and their repeat motifs. Using the buildTRs workflow and 92 haplotype resolved assemblies of taurine and indicine cattle and wisent, we built a repeat definition file for 809,222 TR loci (excluding homopolymers) with loci up to 9.6kb in length. We used this annotation file to genotype TRs/VNTRs with TRGT in 231 long read cattle alignments. Although VNTR alleles are often differentiated by length alone, we used TRGT's cluster algorithm to differentiate alleles by both repeat count and motif variation as both types of VNTR variation had been associated with complex traits and disease in humans. TR/VNTR genotyping of autosomal chromosomes identified 29,011 invariant TRs, 368,593 low variability VNTRs (major allele frequency >95%), and 384,337 VNTR loci (major allele frequency < 95%). To assess how VNTR variation impacts gene expression, we performed cis-eQTL (±1Mb) analyses in 117 individuals that had testis RNA sequencing data. We found significant associations of 3825 genes with mean VNTR allele length. We complemented the mean allele length-based analyses with a mixed model designed to detect unknown features of VNTR allele variation associated with gene expression. Multiallelic VNTR genotypes were encoded into an allele dosage matrix (Z) from which we modeled allele specific covariance across individuals to gene expression. This mixed model approach identified ~4621 genes with significant associations to VNTRs. Although ~3243 of the genes with mean allele length-based associations were also identified by the mixed model, an additional 796 genes found by the mixed model do not have significant mean allele length-based associations. This suggests that additional sources of variation in the VNTR alleles such as motif variability may modulate gene expression. We investigate allele specific variation of VNTRs in cattle and demonstrate how their sequence composition decomposes variation of molecular phenotypes and how VNTRs compare to more traditional neighboring biallelic features (SNPs/INDELs).
Keywords: 2026
How to Cite:
Mason, V., Kadri, N., Mapel, X., Leonard, A. & Pausch, H., (2026) “Assessing the diversity and functional consequences of multiallelic VNTRs in cattle with long-read sequencing”, World Congress on Genetics Applied to Livestock Production Digital Archive 2026(1): 2286464. doi: https://doi.org/10.31274/wcgalp.24018
Rights: 1
Downloads:
Download PDF
View PDF
64 Views
22 Downloads