| Pig is one of China’s most critical agriculture and animal husbandry industries.Although China is one of the largest countries in the world,which has the most pigs and the most significant demand for pork,there are still some problems,such as low breeding efficiency and long-term dependence on imported pig breeding resources.Therefore,it is a significant task for the industry and academia to accelerate the process of pig breeding,select high-quality pig breeds,and reduce the dependence on foreign pig breed resources.The reproduction of pigs is an important economic trait,and improving reproduction could bring higher economic benefits.However,reproduction is a low heritability character,the genetic structure is very complex,and it is challenging to make efficient progress using traditional breeding methods.With the development of molecular biology and breeding techniques in recent years,multi-omics methods such as Genome-wide association study(GWAS)and transcriptomic analysis have been widely used to find candidate genes for traits.Genomic Selection(GS)has been commonly used in pig breeding,significantly improving breeding efficiency.In our study,took 832 Large White sows from a National Pig Core Breeding Farm in Henan Province,focusing on the Total number of piglets born(TNB),Number of piglets born alive(NBA),Mummified piglets(MUM),Litter of born weight(LBW)and Gestation Length(GL).We did the heredity parameter estimate,GWAS,GS,and Weighted gene correlation network analysis(WGCNA)to find the reproduction-related genes.Besides,the Pig Haplotypes Reference Panel(PHARPv2)and Swine Imputation Server(SWIM)were used to fill the chip data to improve the accuracy.Below are key research findings:1.Field,year,season,birth parity,and other non-genetic factors affect reproductive performance.The birth season had a significant effect on the MUM(P<0.05);The field,year,season,and birth parity on reproductive traits of Large White sows were highly significant(P<0.01).Large white sows have the best reproductive performance in spring and good reproductive performance in 3-5 parity.The heritability of TNB,NBA,MUM,and LBW in Large White sows were 0.1638,0.1045,0.0141,and 0.1231,respectively,which belongs to the low heritability traits.The heritability of GL is 0.2194,which belongs to the medium heritability character.The phenotypic correlation and genetic correlation of NBA,TNB,and LBW showed positive and all within the range of 0.68-0.94.There is a negative genetic correlation between GL and NBA,TNB,and LBW.When GL is 115 days or longer,TNB and NBA decrease.The prolongation of GL may be related to the maturation of the sow’s main organs and the development of embryos.2.Mixed Linear Model(MLM),Compress MLM(CMLM),The fixed and random model Cycling Probability Unification(Farm CPU),and Bayesian-information and Linkage-disequilibrium Iteratively Nested Keyway(BLINK)models were used to perform GWAS on TNB,NBA,LBW,GL,and MUM.Our result showed that the BLINK model could detect the variation sites significantly related to reproductive traits(P<0.05).We found 14 SNP locus and detected four associated candidate genes,including PLCB4,PAQR8,NFATC3,and ROR1.3.The results of GWAS imputed based on the PHARP platform detected more significant loci than those imputed with SWIM.Based on the GWAS analysis of two imputation data,identified a total of 57 significant SNPs related to reproductive traits and six candidate genes were discovered,including AKT3,CHST8,NFATC3,LTBP1,SLC24A3,and BYSL.4.The GS accuracy of the imputation dataset trimmed with r~2 of 0.1 is the highest in TNB,NBA,MUM,and LBW traits.Besides,the GS accuracy of the imputation dataset trimmed with r~2 of 0.9 is the highest in the GL trait.Compared with SNP chip data,the accuracy of GS based on the imputation data was improved.5.This study conducted WGCNA based on 23 transcriptome sequencing data from the GEO library and obtained 13 co-expression modules.We found that blue,magenta and turquoise modules strongly correlated with pig reproductive traits in co-expression modules.We detected ten candidate genes through the enrichment analysis and network mapping,including TRMT10A,PRR5L,POLR1D,SUGT1,WDR1,SEC24D,SEMA3C,PTBP1,NOL6,and SLC6A8.GWAS and WGCNA identified an essential candidate gene WDR1,which played a critical role in embryonic heart development,embryo implantation,and birth weight.In summary,this study provides a comprehensive study of reproductive traits in large white sows using different GWAS models and different imputation data.The imputation strategy improved the efficacy of GWAS detection and the accuracy of GS,while key candidate genes affecting reproductive traits were identified in combination with the WGCNA method.This study will help to elucidate the genetic basis of pig reproductive traits and provide an important reference for pig genome breeding. |
