Comparison of estimated breeding values, daughter yield deviations and de-regressed proofs within a whole genome scan for QTL
Autor: | C. Looft, Ingolf Russ, M. Förster, Christa Kühn, G. Averdunk, N. Xu, S. Grupe, Manfred Schwerin, Hauke Thomsen, G. Erhardt, Gudrun A. Brockmann, J. Blümel, R. Reents, F. Reinhardt, Stefan Hiendleder, Ernst Kalm, Norbert Reinsch, I. Medjugorac, B. Leyhe-Horn, Bertram Brenig |
---|---|
Rok vydání: | 2001 |
Předmět: |
0303 health sciences
Daughter Variables media_common.quotation_subject 0402 animal and dairy science food and beverages Genome Scan 04 agricultural and veterinary sciences General Medicine Biology Quantitative trait locus 040201 dairy & animal science Regression 03 medical and health sciences Food Animals Yield (chemistry) Statistics Trait Animal Science and Zoology 030304 developmental biology media_common |
Zdroj: | Journal of Animal Breeding and Genetics. 118:357-370 |
ISSN: | 1439-0388 0931-2668 |
DOI: | 10.1046/j.1439-0388.2001.00302.x |
Popis: | An important issue in quantitative trait loci (QTL) detection is the use of phenotypic measurement as a dependent variable. Daughter yield deviations (DYDs) as the unit of choice are not available for all traits of interest. The use of de-regressed proofs (DRPFs) of estimated breeding values (EBVs) is an alternative to using daughter yield deviations. The objective of this study was to examine possible differences between DYDs and DRPFs within the use of QTL detection. The pedigree used was part of the granddaughter design of the German QTL effort. Consisting marker maps for livestock species were derived from all available data of 16 German Holstein paternal half-sib families with a total of 872 sires. The number of progeny ranged from 19 to 127. A whole genome scan was performed using weighted and unweighted multimarker regression with DYDs, DRPFs and EBVs as dependent variables for the traits milk, fat and protein yields. Results were compared with respect to the number of QTL detected. A similar number of QTL was detected with DRPFs and DYDs. Also, when dependent variables were weighted according to the variance of the trait, a higher number of QTL was detected at the desired level of significance as compared to using unweighted variables. |
Databáze: | OpenAIRE |
Externí odkaz: |