Investigating the impact of reference assembly choice on genomic analyses in a cattle breed

Abstract

BackgroundReference-guided read alignment and variant genotyping are prone to reference allele bias, particularly for samples that are greatly divergent from the reference genome. A Hereford-based assembly is the widely accepted bovine reference genome. Haplotype-resolved genomes that exceed the current bovine reference genome in quality and continuity have been assembled for different breeds of cattle. Using whole genome sequencing data of 161 Brown Swiss cattle, we compared the accuracy of read mapping and sequence variant genotyping as well as downstream genomic analyses between the bovine reference genome (ARS-UCD1.2) and a highly continuous Angus-based assembly (UOA_Angus_1).ResultsRead mapping accuracy did not differ notably between the ARS-UCD1.2 and UOA_Angus_1 assemblies. We discovered 22,744,517 and 22,559,675 high-quality variants from ARS-UCD1.2 and UOA_Angus_1, respectively. The concordance between sequence- and array-called genotypes was high and the number of variants deviating from Hardy-Weinberg proportions was low at segregating sites for both assemblies. More artefactual INDELs were genotyped from UOA_Angus_1 than ARS-UCD1.2 alignments. Using the composite likelihood ratio test, we detected 40 and 33 signatures of selection from ARS-UCD1.2 and UOA_Angus_1, respectively, but the overlap between both assemblies was low. Using the 161 sequenced Brown Swiss cattle as a reference panel, we imputed sequence variant genotypes into a mapping cohort of 30,499 cattle that had microarray-derived genotypes. The accuracy of imputation (Beagle R2) was very high (0.87) for both assemblies. Genome-wide association studies between imputed sequence variant genotypes and six dairy traits as well as stature produced almost identical results from both assemblies.ConclusionsThe ARS-UCD1.2 and UOA_Angus_1 assemblies are suitable for reference-guided genome analyses in Brown Swiss cattle. Although differences in read mapping and genotyping accuracy between both assemblies are negligible, the choice of the reference genome has a large impact on detecting signatures of selection using the composite likelihood ratio test. We developed a workflow that can be adapted and reused to compare the impact of reference genomes on genome analyses in various breeds, populations and species.

Structural variant-based pangenome construction has low sensitivity to variability of haplotype-resolved bovine assemblies

“…More non-reference sequence was identified in the lower-quality Angus-backed pangenome (+13%), while the more complete Simmental-backed pangenomes had less (−6%). Reference-bias propagates through minigraph’s pangenomes, such as the missing sequence in Angus chromosome 28 (Lloret-Villas et al, 2021) resulting in 25% fewer bubbles compared to using ARS-UCD1.2 (Figure 4d).…”

Size and composition of haplotype reference panels impact the accuracy of imputation from low-pass sequencing in cattle

“…Multi-sample variant calling was performed with the GATK HaplotypeCaller, GenomicsDBImport and GenotypeGVCFs modules according to the best practice guidelines (39,40). We applied the VariantFiltration module for site-level filtration with thresholds indicated in (30) to retain high-quality SNP and INDELs.…”

“…Reads were split per read groups with gdc-fastq-splitter (27) (version 1.0.) and subsequently aligned with bwa-mem2 (28) using the -M and -R flags to a manually curated version of the current bovine Hereford-based reference genome (ARS-UCD1.2) (29) that included a Y chromosome as described in (30). Samblaster (31) (version 0.1.26), Sambamba (32), samtools (33,34) (version 1.12), and Picard tools (35) (version 2.25.7) were used to deduplicate and sort the BAM files.…”

A setback into a success: What can batch effects tell us about best practices in genomics?

“…While this shift has merits based on the fact that many population genomics analyses produce more robust results with more samples at low coverage than with fewer samples at high coverage (Alex Buerkle & Gompert, 2013; Lou et al, 2021) remind us that low‐coverage whole‐genome sequencing is more sensitive to artefacts due to DNA degradation, depth heterogeneity or DNA quality. That being said, some artefacts such as reference bias and alignment errors are equally problematic with high‐coverage data (Gage et al, 2019; Lloret‐Villas et al, 2021), and more importantly, Lou and Therkildsen (2022) show that appropriate bioinformatic procedures are key to control and correct for the impact of multiple factors. Most of those mitigation methods include more stringent filtering that may reduce the fraction of genome actually analysed or the number of polymorphic markers.…”