| Growth traits, reproduction traits and carcass traits are three economically important traits in the poultry industry. However, the conventional method of improving product performance by seed selection is costly, both in terms of time and money. Therefore, marker assisted selection (MAS) are used to support breeding programs and become to be an important aspect of breeding work. Two methods are widely used to select genetic molecular markers:candidate genes and QTL mapping, however, both of them have some deficiencies. Currently, GWAS employs SNPs as markers; these potential markers are distributed throughout the entire genome at a high density. Such studies have served to comprehensively analyze complex, economically important traits (such as growth traits, reproduction traits, etc.) with the help of technical statistical tools. Thus, GWAS has been one of the most effective approaches for identifying related QTLs and functional genes. One of the biggest advantage for GWAS is that GWAS do not need an uncertainty hypothesis, it is a direct association study between molecular markers and economic traits.Therefore, a kind of chicken Illumin 60K SNP panel were utilized to analyze ten reproduction traits, twelve growth traits and seventeen carcass and meat quality traits of Jinghai Yellow Chicken. The purpose of this study was to identify key SNPs affecting these important economic traits, discuss distribution characteristics and explore candidate genes. All these results would benefit the molecular researches and provide the good foundation for the breed selection. The main result are shown as follows:1. A total of four models including two linear regression models (LRM Ⅰ and LRMⅡ), a generalized linear model (GLM) and a mixed linear model (MLM) were used to analyze the associations between genotype data and age at first egg (AFE) and body weight at first egg (BWF). Bonferoni correction was used to correct the error caused by multiple comparisons. The result showed that the numbers of significant SNPs identified by four models were different and they had different advantages and disadvantages. Among them, MLM had the most precision, however, it would caused false negative result for some traits. LRM Ⅱ identified more significant SNPs, however, it could cause false positive result for some traits. Therefore, both MLM and LRM Ⅱ models were used in the following analysis.2. The GWAS on ten reproductive traits in 400 Jinghai Yellow Chicken was performed and twenty-seven SNPs reaching genome-wide significance (P<1.8E-06) were found to be associated with bodyweight at first egg (9), egg weight (8), egg numbers (7) and hatchability (3). Twenty-one proximal genes were found and the functions of some particular ones were discussed, such as FAM184B, GPC, KCNIP4, CBFB and STK31, etc., among others. In addition,159 potential significant SNPs (1.8E-06<P<3.59E-05) for ten reproduction traits were also identified. Haplotype analysis identified 5,650 haplotypes in the whole genome (R2>0.8), sixteen haplotypes were identified to reach genome-wide significance with BWF, FEW, EW300 and HA. A few halotypes had new proximal genes. Furthermore, the region 151.3-156.3Mb on GGA1 had the most significant SNPs for bodyweight at first egg.3. In the GWAS of growth traits, LRM Ⅱ identified 25 SNPs reaching genome-wide significance with growth traits (P<1.8E-06). Twenty-one proximal genes were found, such as HTT, SEPT8, FRYL and METTL8, etc. An additional 187 SNPs reaching potential significance with growth traits were also identified (1.8E-06<P<3.59E-05). Moreover, MLM identified nine SNPs reaching potential significance with growth traits (1.8E-06<P<3.59E-05), eight important candidate genes were found and some particular ones were discussed. Most of the SNPs were not reported before, but some of them overlapped the region reported in previous studies. Moreover, a few SNPs were identified to affect more than two growth traits. MAPK and Wnt pathway were found to have affects on bodyweight. Haplotype analysis identified twenty-one significant haplotypes (P<1.8E-06) affecting BW4 (1), BW6 (2), BW12 (2), BW14 (2), BW16 (12) and ADG4 (2). Seven of the twenty-one haplotypes were in 152.4-156.3Mb on GGA1. Moreover, the region 151.3-156.7Mb on GGA1 had 24 significant or potential significant SNPs associated with almost all the growth traits. The main genes in this region were SL1TRK family.4. In the GWAS of carcass and meat quality traits, LRMⅡ identified 27 SNPs reaching genome-wide significance with carcass and meat quality traits (P<1.8E-06). The region 75.5-79.3Mb on GGA4 were identifed to have SNPs affecting multiple carcass traits. Thirteen proximal genes were found, such as GRIK1, CACNA2D2 and FAM184B, etc. An additional 130 SNPs reaching potential significance with growth traits were also identified (1.8E-06<P<3.59E-05). Moreover, MLM identified seventeen SNPs reaching potential significance with growth traits (1.8E-06<P<3.59E-05), thirteen important candidate genes were found, such as FAM184B, LDB2, GFRL1 and CACNA2D2, etc. Seven of the seventeen SNPs were located in 78.7-78.3Mb on GGA4. Moreover, some of the significant or potential significant SNPs were found to affect more than three carcass traits. Haplotype analysis identified fifteen significant haplotypes (P<1.8E-06) affecting BW66 (1), FW (6), WW (2), AW (2), AWP (2) and SEW (1). Nine of the fourteen haplotypes were in 75.5-79.3Mb on GGA4. A few new candidate genes were found and dicussed. |
PDF Full Text Request