Abstract
The whole-genome sequencing (WGS) data can potentially discover all genetic variants. Studies have shown the power of WGS for genome-wide association study (GWAS) lies in the ability to identify quantitative trait loci and nucleotides (QTNs). However, the resequencing of thousands of target individuals is expensive. Genotype imputation is a powerful approach for WGS and to identify causal mutations. This study aimed to evaluate the imputation accuracy from genotyping-by-sequencing (GBS) to WGS in two pig breeds using a resequencing reference population and to detect single-nucleotide polymorphisms (SNPs) and candidate genes for farrowing interval (FI) of different parities using the data before and after imputation for GWAS. Six hundred target pigs, 300 Landrace and 300 Large White pigs, were genotyped by GBS, and 60 reference pigs, 20 Landrace and 40 Large White pigs, were sequenced by whole-genome resequencing. Imputation for pigs was conducted using Beagle software. The average imputation accuracy (allelic R 2) from GBS to WGS was 0.42 for Landrace pigs and 0.45 for Large White pigs. For Landrace pigs (Large White pigs), 4,514,934 (5,533,290) SNPs had an accuracy >0.3, resulting an average accuracy of 0.73 (0.72), and 2,093,778 (2,468,645) SNPs had an accuracy >0.8, resulting an average accuracy of 0.94 (0.93). Association studies with data before and after imputation were performed for FI of different parities in two populations. Before imputation, 18 and 128 significant SNPs were detected for FI in Landrace and Large White pigs, respectively. After imputation, 125 and 27 significant SNPs were identified for dataset with an accuracy >0.3 and 0.8 in Large White pigs, and 113 and 18 SNPs were found among imputed sequence variants. Among these significant SNPs, six top SNPs were detected in both GBS data and imputed WGS data, namely, SSC2: 136127645, SSC5: 103426443, SSC6: 27811226, SSC10: 3609429, SSC14: 15199253, and SSC15: 150297519. Overall, many candidate genes could be involved in FI of different parities in pigs. Although imputation from GBS to WGS data resulted in a low imputation accuracy, association analyses with imputed WGS data were optimized to detect QTNs for complex trait. The obtained results provide new insight into genotype imputation, genetic architecture, and candidate genes for FI of different parities in Landrace and Large White pigs.
Highlights
Reproductive traits play an important role in pig industry and directly affect the sow reproductive performance
For most chromosomes, the accuracy for each chromosome was higher for Large White pigs than for Landrace pigs. This can be explained by there being only 20 Landrace pigs in the reference population compared with 40 Large White pigs. These results demonstrate that the reference population size contributed to imputation accuracy and that imputation accuracy increased with increasing population size
Using a single-breed reference population, imputation from GBS to WGS data resulted in a poor imputation accuracy
Summary
Reproductive traits play an important role in pig industry and directly affect the sow reproductive performance. The researchers extensively studied reproductive traits (such as litter size, birth weight, and number of teats) and identified many quantitative trait loci (QTLs) and candidate genes in pigs. FI was defined as the number of days between two consecutive litters in sow’s productive life. This trait is one of the major determinants of the efficiency of sow reproduction. The previous literatures have shown that these traits in different parities should be considered as different traits for farrowing traits (Roehe and Kennedy, 1995; Noguera et al, 2002; Onteru et al, 2011). 2,129 QTLs were associated with reproductive traits. There are no QTLs and genes found for FI in pigs
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
