| Feed is the largest single cost in pig production,accounting for 65-80% of the total pig production cost.For a long time,improving feed utilization efficiency of breeding pigs has been the focus of breeding workers,and it is directly related to the important economic benefits of the pig industry.In breeding,feed conversion ratio(FCR)and residual feed intake(RFI)are usually used to measure the feed efficiency of pigs.The accuracy of FCR and RFI will directly determine the genetic progress of feed efficiency in pigs.With the popularization of automated electronic measurement equipment,the individual pig feed efficiency measurement can be achieved.Each measuring equipment can collect thousands of pig feed data(including feed intake and weight data)on average in one day,but its phenotypic accuracy has been widely questioned by researchers.The main reason is that pig body weight data is relatively easy to obtain,but how to accurately measure pig intake data has been restricting the accuracy of feed efficiency related traits.Therefore,it is still difficult to identify the quantitative trait loci(QTLs)and key genes related to feed efficiency.In the early stage of this study,9667 Wens S21 and 7068 Wens S22 Duroc sows were collected by using the Osborn production performance measurement system.The measurement period was 7 years(2012-2018).Through the preliminary quality control of the data,it was found that there were obvious systematic errors in pig feeding data,which were as follows:(1)significant outliers or systematic stratification of body weight data caused by machine failure;(2)the amount,time and rate of feeding violated the basic biological knowledge.These factors greatly restrict the progress of genetic improvement of feed efficiency related traits.Based on this,the purpose of this study is to carry out the following panoramic research work on feed efficiency traits:(1)establish a phenotypic refinement processing process suitable for the feeding data of the Osborn production performance measurement system,and estimate the genetic parameters;(2)integrate the genome re sequencing and gene chip data,and identify the Major QTLs affecting feed efficiency related traits of pigs by GWAS,and pay attention to focused on pleiotropic QTLs in two Duroc populations;(3)The copy number variation regions(CNVRs)of two Duroc populations were revealed,CNVRs,which was significantly correlated with feed efficiency,was identified by association analysis,and then provide a valuable reference for the genetic improvement of later traits.The specific conclusions of this study are as follows:(1)Through the quality control,identification,correction and segmented calculation of abnormal ingestion data,the phenotypic values of daily feed intake(DFI),average daily gain(ADG),time spent to eat per day(TPD),number of visits to feeder per day(NVD),time spent to eat per visit(TPV),feed intake per visit(FPV),feed intake rate(FR),FCR and RFI of3,749 S21 Duroc and 4,295 S22 Duroc were obtained.Through further estimation of genetic parameters,it is found that except for FCR and ADG with low heritability(0.10-0.18),the other seven traits are all medium to high heritability(0.31-0.62),which to some extent shows that the refined processing flow of this study for the data of food intake has better effect.It is worth mentioning that the phenotypic variation coefficients of NVD,TPV,FPV and FR are all greater than 20%,suggesting that in the breeding process,the traits of feeding behavior are likely to not undergo artificial selection and have better genetic improvement space.In addition,through the study of large samples,this study also found for the first time that the length of sunshine in different seasons may directly lead to changes in the peak time of pig feeding,and the relevant genetic mechanism needs further study.(2)This study used 50 K SNPs chips to genotype 1,170 S21 and 1,265 S22 Duroc.The molecular pedigree and genetic distance were used to select 152 S21 and 151 S22 Duroc for re-sequencing(10 × ～ 20 × genome).Principal Component Analysis(PCA)shows that although S21 and S22 belong to Duroc,the genetic differences between the lines are much larger than the genetic differences within the lines,suggesting that line factors should not be ignored in subsequent genetic analysis.Then,the Minimac4 software was used to impute the SNPs and In/Del identified by resequencing into 50 K SNPs chip individuals.After quality control,a total of 7,428,052 SNPs and 497,420 In/Del were finally obtained in S21;and in S22,a total of 10,731,999 SNPs and 832,495 In/Del.And the average imputation accuracy of S21 and S22 is as high as 97.6% and 98.4% respectively.Genome wide association study(GWAS)analysis of different lines showed that in S21,1,1,1,5,3,2 and 2 QTLs were detected with FPV,TPV,TPD,ADG,DFI,FCR and RFI were significantly correlated(P<5E-6);in S22,1,1,5,5,1,and 10 QTLs were detected to be significantly correlated with FR,TPD,ADG,DFI,FCR,RFI(P < 5E-6).It is worth mentioning that in the S21 and S22 populations,there is a pleiotropic QTL(1: 160.1Mb-161.8Mb)that significantly affects ADG,DFI,RFI and TPD on chromosome 1 and it is the most significant Associated loci can explain5.62% phenotypic variation of DFI traits(P = 8.54E-13).Linkage disequilibrium analysis showed that MC4 R and CCBE1 genes in this region may be important candidate genes that affect feeding behavior and feed efficiency traits.In addition,in the S22 population,there is also a pleiotropic QTL(1: 34.5Mb-38.5Mb)that affects both DFI and RFI,the most significant association site can explain the 5.44% phenotypic variation of DFI trait(P =1.31E-09),in which HEY2 and NKAIN2 genes are the candidate genes most likely to affect related traits.(3)Use Penn CNV software to detect copy number variation and copy number variation region based on 50 K SNPs chip,and construct a copy number variation map of Duroc population.In S21,3,047 CNVs and 433 CNVRs were detected;in S22,1,816 CNVs and286 CNVRs were detected.Based on the association analysis of CNVRs,the results show that in S21,there are 3,3,1,4,3,4,5,1,and 1 CNVRs and FCR,ADG,DFI,NVD,TPD,TPV,FPV,FR and RFI were significantly correlated(P <0.05);in S21,there were 1,3,2,2,1,1,1,and 2 CNVRs and FCR,ADG,DFI,NVD,TPD,TPV,FPV significantly correlated with RFI(P <0.05).It is worth mentioning that in both populations,CNVR＿72(3: 1,13Mb-1.47Mb)located on pig chromosome 3 affects the NVD,FCR,FPV,and TPD of pigs at the same time.This area contains a total of ELFN1 and MAD1L1.The genetic mechanism of two candidate genes needs further study.This study revealed the genetic mechanism of SNPs,In/Del,and CNVs on feeding behavior and feed efficiency traits,and advanced our understanding of the genetic complexity of pig feeding behaviors and feed efficiency traits.In this study,some SNPs,In/Del and CNVs with "pleiotropic" on feeding behavior and feed efficiency were identified in two Duroc populations.These molecular markers will be directly applied to the molecular genetic improvement of the feed efficiency traits of Duroc pigs,thus laying the foundation for the cultivation of diet-saving Duroc pigs. |
PDF Full Text Request