HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Full Text Full Text (PDF) All Versions of this Article: jas.2008-0852v1 87/3/868    most recent Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Hassen, A. Articles by Dekkers, J. C. M. Search for Related Content PubMed PubMed Citation Articles by Hassen, A. Articles by Dekkers, J. C. M.

The effect of heritability estimates on high-density single nucleotide polymorphism analyses with related animals¹

A. Hassen^*, S. Avendano, W. G. Hill, R. L. Fernando^*, S. J. Lamont^* and J. C. M. Dekkers^*,2

^* Department of Animal Science and Center for Integrated Animal Genomics, Iowa State University, Ames 50011; and Aviagen Ltd., Newbridge, Midlothian, EH28 8SZ, United Kingdom; and Institute of Evolutionary Biology, University of Edinburgh, Edinburgh, EH9 3JR, United Kingdom

² Corresponding author: jdekkers{at}iastate.edu

Analysis of high-density SNP data in outbred populations to identify SNP that are associated with a quantitative trait requires efficient ways to handle large volumes of data and analyses. When using mixed animal models to account for polygenic effects and relationships, genetic parameters are not known with certainty, but must be chosen to ensure proper evaluation of SNP across chromosomes and lines or breeds. The objectives of this study were to evaluate the influence of heritability on the estimates and significance of SNP effects, to develop efficient computational strategies for analysis of high-density SNP data with uncertain heritability estimates, and to develop strategies to combine SNP test results across lines or breeds. Data included sire SNP genotypes and mean progeny performance from 2 commercial broiler breeding lines. Association analyses were done by fitting each SNP separately as a fixed effect in an animal model, using a range of heritabilities. The heritability used had a limited impact on SNP effect estimates, but affected the SE of estimates and levels of significance. The shape of the frequency distribution of P-values for the test of SNP effects changed from a highly skewed L-shaped curve at low heritability to a right-skewed distribution at high heritability. The P-values for alternative heritabilities could, however, be derived without reanalysis based on a strong linear relationship (R² = 0.99) between differences in log-likelihood values of models with and without the SNP at different levels of heritabilities. With uncertain estimates of heritability, line-specific heritabilities that ensure proper evaluation of SNP effects across lines were determined by analysis of simulated sire genotypes and by permutation tests. Resulting heritability estimates were between those obtained from the entire breeding populations and those obtained from the data included in the sample data set. In conclusion, the uncertainty of heritability estimates has a limited impact on SNP effect estimates in association analyses, but a large impact on significance tests. The impact of heritability on tests can, however, be dealt with in a computationally efficient manner by using the strong linear relationship between model statistics under alternate levels of heritability. These approaches allow efficient analysis of large numbers of SNP for multiple traits and populations and pooling of results across populations.

Key Words: association test • high-density single nucleotide polymorphism • linkage disequilibrium mapping