Effect of Discrepancy Between Assumed and Actual Values of Full-sib Resemblance on the Relative Accuracy of Combined Selection
スポンサーリンク
概要
- 論文の詳細を見る
LUSH (1947) discussed the accuracy of combined selection relative to that of mass selection on the assumption that the value of full-sib resemblance (intra-class correlation) used for calculating weights to be given to individual's own phenotype and family average perfectly agrees with the actual value of it in the population to which combined selection is applied. This assumption, however, could not be strictly real in most cases.In the present note, a formula for the relative accuracy of combined selection, in which both the assumed and actual values of full-sib resemblance can vary, was derived to examine to what extent the relative accuracy of combined selection is lowered by discrepancy between such two kinds of values. Numerical examples were given for the cases of family size being four and eight.General conclusions are: 1) As far as the actual value of full-sib resemblance (τ) is around 0.3 or less, combined selection is never less accurate than mass selection, whatever value between 0 and 0.5 may be used as the assumed value (t). 2) Combined selection may become less accurate than mass selection when τ is about 0.3 or larger and t is smaller than 0.3, depending upon the amount of discrepancy between τ and t. 3) At a given level of τ between 0.1 and 0.4, discrepancy of the order of 0.1 in either direction between τ and t lowers accuracy of combined selection only by around 2 per cent compared to the maximum accuracy expected when t equals τ.
- 日本家禽学会の論文
日本家禽学会 | 論文
- ニワトリ MHC クラスIcDNA プローブを用いたニホンウズラゲノミック DNA の ECL 法による RFLP 分析
- 各種家禽の赤血球中のカルノシンおよびアンセリンの検出〔英文〕
- 鶏の加齢と血液中の遊離アミノ酸および低分子ペプチド組成との関係
- ニワトリヒナのファブリシウス嚢におけるグルココルチコイド誘導性のアポトーシスに関する免疫組織化学的検討
- 鶏の体脂肪蓄積に及ぼす分離大豆タンパク質及びカゼインの影響