ARS | Publication request: Use of haplotypes to predict selection limits and Mendelian sampling

Submitted to: Journal of Dairy Science
Publication Type: Abstract Only
Publication Acceptance Date: March 9, 2010
Publication Date: June 24, 2010
Citation: Cole, J.B. 2010. Use of haplotypes to predict selection limits and Mendelian sampling. Journal of Dairy Science. 93(E-Suppl. 1):535(abstr. 623).

Technical Abstract: Limits to selection and Mendelian sampling terms can be calculated using haplotypes, which are sums of individual additive effects on a chromosome. Haplotypes were imputed for 43,385 actual markers of 3,765 Jerseys using the Fortran program findhap.f90, which combines population and pedigree haplotyping methods. Longer chromosomes had more distinct haplotypes, ranging from 7,287 for Bos taurus autosome 1 (BTA) to 2,460 for the X chromosome. This is expected because longer chromosomes undergo recombination more often than shorter ones. Mendelian sampling (MS) variances were calculated for genotyped animals as the sum of squared haplotype differences for each chromosome in the genome. The distribution of MS variances had a heavy right tail (skewness = 0.276), with a mean of 49,290 ± 13,981. Genotypes for each chromosome were constructed from pairwise combinations among the top 5% of haplotypes based on the sum of marker effects for lifetime net merit (NM$) for each chromosome. Correlations among raw and adjusted values in the top group ranged from 0.897 on BTA12 to 0.998 on the X chromosome. Selection of the best unadjusted haplotypes for each chromosome results in an animal with an EBV of +$5,243 for NM$. Adjusting for inbreeding resulted in a slightly lower EBV of +$4,496. Haplotype values were adjusted to account for changes in homozygosity by adding or subtracting 6% of an additive genetic standard deviation per 1% decrease or increase in homozygosity. The top Jersey bull, ALL LYNNS RESTORE VERNON-ET (29JE03647), had an EBV NM$ of +$1,180 in the January 2010 evaluation. For 11 chromosomes (BTA 4, 9, 13, 15, 20, 21, 22, 25, 26, 28, and X) the best genotype after adjusting for inbreeding consisted of two copies of the same haplotype. Differences between the best and poorest haplotypes ranged from a maximum of $65 for BTA1 to a minimum of $12 for BTAX. Selecting animals rather than chromosomes may result in slower progress, but limits may be the same because most chromosomes will become homozygous with either strategy. Selection on functions of MS could be used to change variances in later generations.