Rare and population-specific functional variation across pig lines

Abstract

BackgroundIt is expected that missense and loss-of-function (LOF) variants are responsible for phenotypic differences among breeds, genetic lines and varieties of livestock and crop species that have undergone diverse selection histories. However, there is still limited knowledge about the existing missense and LOF variation in livestock commercial populations, in particular regarding population-specific variation.MethodsWe performed whole-genome re-sequencing of 7,848 individuals from nine commercial pig breeding lines (average coverage: 4.1x) and imputed genotypes for 440,610 pedigree-related individuals. The variants were categorized according to predicted functional annotation (from loss-of-function to intergenic) and prevalence level (number of lines in which the variant segregated; from private to widespread). Variants in each category were examined in terms of distribution along the genome, minor allele frequency, Wright’s fixation index (FST), individual load, and association to production traits.ResultsOf the 46 million called variants, 28% were private (i.e., called in only one line) and 21% were widespread (i.e., called in all nine lines). Genomic regions with low recombination rate were enriched with private variants. Low-prevalence variants were enriched for lower allele frequencies, lower FST, and putatively functional and regulatory variants (including loss-of-function and deleterious missense). Only a small subset of low-prevalence variants was found at intermediate allele frequencies and had large estimated effects on production traits. Individuals on average carried less private deleterious missense alleles than expected compared to other predicted consequence types. A small subset of low-prevalence variants with intermediate allele frequencies and higher FST were detected as significantly associated to the production traits and explained small fractions of phenotypic variance (up to 3.2%). These associations were tagged by other more widespread variants, including intergenic variants.ConclusionsMost low-prevalence variants are kept at very low allele frequency and only a small subset contributed detectable fractions of phenotypic variance. Low-prevalence variants are therefore unlikely to hinder across-breed analyses, in particular when predicting genomic breeding values using reference populations of a different genetic background.

Genomic prediction with whole-genome sequence data in intensely selected pig lines

“…In our multi-line scenarios we only used variation that segregated across all seven lines. We observed that population-specific variation accounted only for small fractions of genetic variance [53] and it seems unlikely that they would contribute much to prediction accuracy across breeds. Another possible obstacle is the differences in the allele substitution effects of the causal mutations across breeds.…”

“…When functional annotation is not considered, intergenic variants are more likely to be preselected by chance. Such variants tend to be more common and widespread across populations, and therefore can act as tag variants and capture much larger fractions of trait variance [53].…”

“…This amounted to a total of 55.6 million SNP (19.6 to 31.1 million within each line) and 10.2 million indels (4.1 to 5.6 million within each line). A more complete description of the variation across the lines is provided in [53].…”