A comparison of restricted selection index and linear programming in sire selection

1992 ◽  
Vol 84-84 (3-4) ◽  
pp. 384-389 ◽  
Author(s):  
T. R. Famula
Genome ◽  
1987 ◽  
Vol 29 (1) ◽  
pp. 91-96 ◽  
Author(s):  
J. L. Campo ◽  
B. Villanueva

A selection index (I) method was compared with independent culling levels (N), with a restriction in the selection program, using two replicated single generation experiments in Tribolium castaneum, which are considered forms of antagonistic selection. The first experiment was designed to increase adult weight without changing pupal weight, while the second experiment was intended to improve egg laying without changing adult weight. In both experiments the genetic correlation between the traits involved were positive but were higher in experiment 1 than in experiment 2. The proportion of selection was 10%. In experiment 1, the effect of restriction was as expected in both lines since the changes in pupal weight were not significant. Adult weight change was positive for the I line and negative for the N line, showing that the index was a superior method to improve adult weight. In experiment 2, the effect of restriction was also as expected in both lines and changes in adult weight were not significant. Egg laying changed positively in both lines. Therefore, both selection methods were similar in this experiment, even though egg laying change was higher in the I than the N line. Key words: restricted index, restricted culling levels, antagonistic selection, Tribolium.


2021 ◽  
Vol 10 (3) ◽  
pp. 202-207

The aim of this investigation was to develop restricted selection index aiming to improve 305-day yields of milk (MY), fat (FY) and protein (PY), while keeping the deterioration in days open (DO), calving interval (CI) and number of services per conception (NSPC) at minimum levels in Holstein cows. The data represent 3682 records of 1122 cows, daughters of 95 sires and 712 dams. The data were analyzed by multi-trait animal model with repeated measures. Eight selection indexes (five unrestricted and three restricted) were derived using MY, FY, PY, CI, DO and NSPC in various combinations as sources of information in the indexes. However, the true breeding value included MY, FY and PY. The highest accuracy of selection (0.60) resulted from selection based on the full index. Milk yield and SPC appeared to be the most valuable traits in the full index. Combining the two traits into one index (the best reduced index) gave 0.57 accuracy of selection. The index based on MY alone (the most accurate single trait index) gave 0.53 accuracy. It seems possible to reduce the expected genetic deterioration in the reproductive traits by restricting the full index to result in zero genetic change in NSPC (rTI=0.48). This restriction will allow the breeder to mitigate the deterioration in DO and CI by 12 and 16 days respectively, by sacrificing with part of the expected genetic improvement in productive traits (29, 40 and 48% in MY, FY and PY, respectively).


1969 ◽  
Vol 6 (4) ◽  
pp. 204-208
Author(s):  
TAKEO ABE ◽  
YUKIO YAMADA ◽  
AKIRA NISHIDA

Sign in / Sign up

Export Citation Format

Share Document