Untangling the Copy Number Variation at the Basis of Belted Phenotypes in Cattle Using Long-Read Sequencing

The belted phenotype in cattle, a distinctive white belt across the midsection observed in breeds such as the Dutch Belted and Belted Galloway, is caused by a copy number variant (CNV) upstream of TWIST2, a gene involved in melanocyte development. While the CNV is necessary for belt formation, the genetic basis of variation in belt width and completeness remains unclear. To refine understanding of these phenotypes, we generated a de novo genome assembly of Dutch Belted cattle and analysed both short- and long-read sequencing data from animals with normal-width, small-width, and incomplete belts. Standard variant calling tools failed to identify the CNV, highlighting technical challenges in detecting large insertions. Instead, coverage analysis and qPCR confirmed the CNV as a 6 kb tandem array approximately 16.5 kb upstream of TWIST2. Long-read evidence showed identical four-copy haplotypes for individuals with normal-width, small-width, and incomplete belts, indicating that CNV copy number does not influence belt width or completeness. Furthermore, no unique variants were detected within the CNV or TWIST2 in a small-width belted individual, suggesting that modifiers affecting belt width likely reside elsewhere in the genome. Genome-wide scans revealed extremely low nucleotide diversity in the CNV region, consistent with a selective sweep driven by strong artificial selection unique to the Dutch Belted breed. Together, these results show that while the CNV is necessary for belt formation, additional modifier loci likely influence belt width and completeness.