Bovine breed-specific augmented reference graphs facilitate accurate sequence read mapping and unbiased variant...

Area of Science:

• Genomics
• Bioinformatics
• Animal Genetics

Background:

• The current bovine genomic reference sequence, derived from a single Hereford cow, exhibits limited diversity and suffers from reference allele bias due to the absence of allelic variation.
• Cattle, with their high nucleotide diversity and numerous breeds, present an ideal model for developing improved, variation-aware reference sequences.

Purpose of the Study:

• To construct and evaluate variation-aware reference graphs for cattle, aiming to overcome the limitations of the existing linear reference sequence.
• To assess the impact of these novel reference graphs on the accuracy of sequence read mapping and variant genotyping.

Main Methods:

• Augmented the existing bovine linear reference (ARS-UCD1.2) with filtered variants from various cattle breeds (dairy and dual-purpose) using the vg toolkit.
• Constructed breed-specific and pan-genome reference graphs, including a whole-genome graph incorporating 14 million alleles from the Brown Swiss breed.
• Evaluated mapping accuracy using pre-selected variants versus random variants in graph construction.

Main Results:

• Read mapping accuracy significantly improved when using variation-aware reference graphs constructed with pre-selected variants compared to the linear reference.
• Both breed-specific and pan-genome graphs demonstrated comparable improvements in mapping accuracy over the linear reference.
• The developed whole-genome graph facilitated accurate read mapping and unbiased genotyping of single nucleotide polymorphisms (SNPs) and insertions/deletions (Indels).

Conclusions:

• Developed the first variation-aware reference graph for an agricultural animal, enhancing sequence read mapping and variant genotyping.
• This novel reference structure represents a significant advancement over linear references for species with substantial genetic diversity.
• This work paves the way for transitioning to variation-aware reference structures in other species with complex population genetics.