Premium
Effects of selection on early stage genetic evaluation for berry weight in breeding table grapes
Author(s) -
Wei X.,
Sykes S. R.,
Clingeleffer P. R.
Publication year - 2003
Publication title -
plant breeding
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.583
H-Index - 71
eISSN - 1439-0523
pISSN - 0179-9541
DOI - 10.1046/j.1439-0523.2003.00757.x
Subject(s) - heritability , berry , biology , selection (genetic algorithm) , table grape , table (database) , bivariate analysis , population , statistics , horticulture , mathematics , demography , genetics , artificial intelligence , sociology , computer science , data mining
Abstract The effects of random and directional selection on estimates of heritability and estimated breeding values were investigated using data collected for grape berry weight during seasons 1999 and 2000 from the CSIRO table grape breeding population. In 1999, all berry‐bearing vines available (4834) were assessed, while in 2000 a randomly selected sample of 1344 vines was evaluated. In addition, a simulated directional selection was imposed such that the berry weights for half of the vines with lighter berries sampled in 2000 were discarded to produce a biased data set. The results showed that heritability of berry weight in 2000 was biased downward and estimated breeding values were over‐ or under‐estimated if analyses were based on directionally selected data sets alone. By incorporating data from 1999 into bivariate analyses, bias in heritability and EBV could be eliminated to a large extent. The implications for grapevine breeding are discussed.