P5041 Searching for allelic distortion in RNA-seq data from boar’s mature sperm

Abstract

Boars have been historically selected for meat and carcass quality. However, breeders are starting to pay attention to additional traits that could be included as selection goals. This is the case for sperm quality, as up to 30% of the boars that enter insemination centers are rejected due to poor values on sperm phenotypes. We have previously performed RNA-seq using mature sperm samples from 6 adult boars and profiled their transcriptome. The objective of the present study was to analyze another layer of information of this RNA-seq experiment by calling single nucleotide polymorphism (SNPs) and Insertion/Deletions (INDELs) and searching for statistically significant allelic distortions in sperm relevant genes. RNA libraries were prepared with two different kits to compare performances: (1) SMARTer Universal Low Input RNA (Clontech) and (2) TruSeq (Illumina), sequenced on an Illumina HiSeq 2000 platform. RNA-seq reads were mapped onto the swine genome (ssc10.2) with TopHat 2. Variants were called with SAMtools/BCFtools and filtered with VCFtools. Then, the Variant Effect Predictor tool (VEP) was used to predict the effect of the coding variants on the affected protein. The low amount and quality of RNA molecules in mature sperm was reflected in the RNA-seq mapping data. Both the SMARTer and the Truseq sequencing reads poorly mapped to the swine genome. For SNP calling, we used a total of 804,328 and 141,288 reads for the SMARTer and the TruSeq, respectively. Preliminary results showed better performance of the TruSeq for SNP/INDEL calling, with an average of 393 SNPs/INDELs per sample called with a minimum read depth of 5, compared to 156 reads obtained with the SMARTer. To assess potential allelic distortion, we screened SNPs with a read depth above 10, allelic ratios (number of reads with reference alleles/alternative allele reads) < 0.4 or > 0.6 and a Chi-squared Test with a p-value < 0.05. Results showed some SNPs with significant allelic read differences in at least one pig in genes related to spermatogenesis and sperm motility as for example the ODF1 gene (Outer Dense Fiber Of Sperm Tails 1). Our preliminary results suggest the utility of RNA-seq to identify SNPs potentially relevant to sperm quality. RNA-seq with higher sequencing depth, in more samples and combined with genome-wide association scan are needed to better exploit allelic distortion analysis in the search for DNA markers of sperm quality in boars.