چكيده انگليسي :
Examining the effective factors on genetic evaluation of various traits is important in animal breeding and poultry industry. The purposes of the present study are to survey population structure and genetic diversity, comparing the accuracy and bias of the estimated breeding values (EBV) using BLUP methods consisted on pedigree (BLUP) and single-step genomic BLUP (ssGBLUP). Five levels of allelic frequency were performed and genomic breeding values for body weight (BW) and average daily gain (ADG) traits, were evaluated. Data were from records of 488 F2 chickens (312 birds were genotyped and 176 birds not genotyped) resulting from cross-breeding of fast-growing Arian chickens and slow-growing Urmmia native chickens from ages of 2 to 7 weeks for BW and 2 to 4 weeks for ADG. Genetic structure analysis of the population by MDS, heat map and neighbor-joining phylogenic tree confirmed the grouping of 312 F2 genotyped birds in 8 subpopulations. The average expected heterozygosity in different groups were estimated as K1=0.4677, K2=0.4635, K3=0.4949, K4=0.5235, K5=0.4613, K6=0.4273, K7=0.4618 and K8=0.4459. Also, the Fixation index (Fst) in these subgroups varied from 0.01 to 0.29. After passing the quality control stage, 48379 SNPs were divided into five subgroups with different minor allele frequencies (MAF) of 0.05-0.1, 0.1-0.2, 0.2-0.3, 0.3-0.4 and 0.4-0.5. Average accuracy of genomic prediction in ssGBLUP compared to the BLUP methodology for BW using the 5-fold cross validation (CV) method in ages of 2, 3, 4, 5, 6 and 7 weeks were improved by 59.03%, 220.37%, 0.46%, 5.61%, 0.45% and 2.73%, respectively. Also, in comparison to using 100% of SNPs, using subset MAFs of 0.05-0.1, 0.1-0.2 and 0.4-0.5 0.6%, 5.4% and 0.6%, in the second week, MAF of 0.4-0.5 16.66% in the third week, MAFs of 0.3-0.4 and 0.4-0.5 8.37% and 6.05% in the fourth week, MAFs of 0.3-0.4 and 0.4-0.5 3.94% and 4.5% in the fifth week, MAFs of 0.1-0.2 and 0.4-0.5 7.31% and 4.11% in the sixth week, and MAFs of 0.3-0.4 and 0.4-0.5 7.54% and 8.91% in the seventh week genetic improvement for prediction accuracy were obtained for BW trait, respectively. Moreover, for ADG, the average accuracy of genomic prediction using ssGBLUP improved comparing to BLUP in weeks 2, 3 and 4, by values 1.96%, 3.87% and 2.12% respectively. For the average accuracy of genomic prediction in the ssGBLUP method using different MAF subgroups in comparison to using all SNPs, MAFs of 0.3-0.4 and 0.4-0.5 6.86% and 0.98% in week 2, MAFs of 0.3-0.4 and 0.4-0.5 1.93% and 8.38% in week 3, MAFs 0.1-0.2, 0.2-0.3, 0.3-0.4 and 0.4-0.5 8.51%, 3.19%, 11.7% and 12.77% improvement in week 4 were achieved, respectively. In addition, the superiority of using subgroup of SNPs with MAF of 0.4-0.5 compared to the full set of SNPs improvs the accuracy of genomic prediction for BW and ADG in different weeks. In general, the results showed that the use of markers with allelic frequency of 0.4-0.5 can be used to reliably rank individuals based on the genetic merit of growth related traits.
