On the use of the transmission disequilibrium test to detect pseudo-autosomal variants affecting traits with sex-limited expression

We herein describe the realization of a genome-wide association study for scrotal hernia and cryptorchidism in Norwegian and Belgian commercial pig populations. We have used the transmission disequilibrium test to avoid spurious associations due to population stratification. By doing so, we obtained genome-wide significant signals for both diseases with SNPs located in the pseudo-autosomal region in the vicinity of the pseudo-autosomal boundary. By further analyzing these signals, we demonstrate that the observed transmission disequilibria are artifactual. We determine that transmission bias at pseudo-autosomal markers will occur (i) when analyzing traits with sex-limited expression and (ii) when the allelic frequencies at the marker locus differ between X and Y chromosomes. We show that the bias is due to the fact that (i) sires will preferentially transmit the allele enriched on the Y (respectively X) chromosome to affected sons (respectively daughters) and (ii) dams will appear to preferentially transmit the allele enriched on the Y (respectively X) to affected sons (respectively daughters), as offspring inheriting the other allele are more likely to be non-informative. We define the conditions to mitigate these issues, namely by (i) extracting information from maternal meiosis only and (ii) ignoring trios for which sire and dam have the same heterozygous genotype. We show that by applying these rules to scrotal hernia and cryptorchidism, the pseudo-autosomal signals disappear, confirming their spurious nature.