Comprehensive detection of structural variations in long and short reads dataset of French cattle

Structural variants (SVs) correspond to different types of genomic variants larger than 50 bp. Many findings suggest the use of long-read (LR) rather than short-read (SR) sequencing to improve the accuracy of SVs detection. Here, we present the results of an in-depth analysis for detection of SVs, mainly large insertions and deletions, in 14 French bovine breeds, based on whole-genome sequence (WGS) data comprising 176 LR and 571 SR samples, with 154 individuals having both LR and SR data available. We first investigated possible biases on the performances of well-known SVs detection tools, namely CUTESV, PBSV, and SNIFFLES, using LR from different technologies, including PacBio HiFi, Oxford ONT, and PacBio CLR. We subsequently highlighted the abilities of tools for detecting SVs (DELLY, LUMPY, and MANTA) and for genotyping known SVs (GRAPHTYPER, SVTYPER, PARAGRAPH, and VG toolkit) using SR data. We then show how the incremental composition of samples in the reference panel affected the SVs genotyping for six validation individuals sequenced in SR. We then searched for the optimal parameters and created the final SVs reference panel consisting of 25,191 deletions and 30,118 insertions. Finally, we emphasized the landscape of the genotyped SVs segregating across 571 SR individuals of 14 breeds.