Presentation is loading. Please wait.

Presentation is loading. Please wait.

Using Haplotypes in Breeding Programs

Similar presentations


Presentation on theme: "Using Haplotypes in Breeding Programs"— Presentation transcript:

1 Using Haplotypes in Breeding Programs

2 What is imputation? Prediction of unknown genotypes from observed genotypes Pedigree haplotyping Matching allele patterns Genotypes indicate how many copies of each allele were inherited Haplotypes indicate which alleles are on which chromosome

3 Why impute haplotypes? Predict dam from progeny SNP
Predict unknown SNP from known Measure 3,000, predict 50,000 SNP Measure 50,000, predict 850,000 Measure each haplotype at highest density only a few times Increase reliabilities

4 How does imputation work?
Identify haplotypes in population using many markers Track haplotypes with fewer markers e.g., track a 20 SNP block with 4 SNP 4 SNP: 2202 20 SNP:

5 How well does haplotyping work?
Half of young animals assigned 3K Proven bulls, cows all had 50K Dams imputed using 50K and 3K Half of ALL animals assigned 3K Could 3K CDDR genotypes help? 10,000 proven bulls yet to genotype Should cows with 3K be predictors?

6 Correlations of 3K and PA with 50K
Half of young animals had 3K PTA, half had 50K PTA Trait Corr(3K,50K)2 Corr(PA,50K)2 Gain NM$ .899 .518 79% Milk .920 .523 83% Fat .516 Prot .555 82% PL .933 .498 87% SCS .912 .417 85% DPR .937 .539 86%

7 Using 3K as reference genotypes
Half of ALL animal NM$ were from 3K, half 50K REL gain as compared to all 50K Breed 50K 3K Imputed dams HO 90% 73% 36% JE 82% 56% 44% BS 84% 72% 55%

8 What about the HD chip? In simulation, reliability averaged 84.0% for 500K SNP versus 82.6% for 50K SNP Haplotypes can fill in almost all missing SNP when moving from 50K to 500K

9 Selection limits and Mendelian sampling
Genotypes for 43,385 SNP using the Illumina BovineSNP50 Haplotypes imputed with findhap.f90 (VanRaden et al., 2010) 1,455 Brown Swiss bulls and cows 40,351 Holstein bulls and cows 4,064 Jersey bulls and cows Results shown for net merit

10 How far can we go with selection?
Breed Best chromosomes Best individual SNP BS 3,857 4,569 HO 7,526 11,841 JE 4,678 5,758

11 How much sampling variance is there?
Breed All loci unlinked Perfect linkage BS 2,359 40,458 HO 21,581 94,343 JE 3,978 44,552

12 Why are HO different from BS and JE?
BS & JE: No QTL HO: Multiple QTL for NM$ ― DGAT1 on BTA 14, AIPL marker on BTA 18

13 What can we conclude? Higher densities probably less valuable than additional animals Imputation can accurately fill in missing SNP when moving from lower to higher densities Selection limits and sampling variances differ among breeds


Download ppt "Using Haplotypes in Breeding Programs"

Similar presentations


Ads by Google