[ Pobierz całość w formacie PDF ]

nr
P' +________(P-P').
1+(n-1)r
13 All domestic livestock have several traits of economic importance,
and their relationships to each other are critical for selection
programs. A correlation describes the relationship between two traits.
There are phenotypic, genetic, and environmental correlations. A
correlation can have values between -1.0 and +1.0, with zero meaning no
relationship. The nearer the correlation is to +1.0 or -1.0, the closer
the relationship is between the two traits. A positive correlation (+)
indicates high measures of one trait tend to occur with high values of
the second and low values for the first with low values for the second.
A negative correlation (-) indicates a tendency for high values of one
trait to occur with low values of the second. For example, milk and fat
yields of goats are positively correlated. As pounds of milk per
lactation increase so do the pounds of fat produced. However, milk
yield and milk fat percentage are negatively correlated. As pounds of
milk increase, the percentage of fat in the milk of goats tends to
decrease.
14 Genetic correlations are important in selection and have two
biological causes: pleiotropy and linkage. Pleiotropy is the result of
one gene contributing to the phenotype of more than one trait. Linkage
means a gene (or set of genes) is in close proximity on a chromosome to
a gene for a second trait. Being close together on the same chromosome,
http://www.inform.umd.edu/EdRes/Topic/AgrEn...dd/goat/SELECTION_AND_BREEDING_SYSTEMS.html (3 of 10) [4/15/2004 12:34:25 PM]
SELECTION AND BREEDING SYSTEMS
they are passed on to the progeny together and cause the genetic
correlation. Thus selection for one trait will alter also the performance
of the population for all other traits which are genetically correlated
to the trait under direct selection. That change in a correlated trait
is called a correlated response. Some correlated responses can be
beneficial in terms of improving the total productivity of goats;
however,
others may be detrimental. Genetic correlations as well as phenotypic
correlations may be used in indexing animals for simultaneous selection
of more than one trait.
15 Selection Response
The first step in the selection process is to define the goals of the
program, e.g., which trait or traits are desired in selection. The
appropriate records need to be collected on the selection candidates and
their relatives. From these records, the BV's of the individuals are
estimated and the goats ranked from best to worst. The breeder must now
decide how many goats are needed for both sexes, and selection is then
simply keeping the top ranked animals. Fewer bucks are required to
maintain the population than females, therefore the intensity of
selection for males can be much greater. This points out that more
progress can be made by concentrating efforts on buck selection.
16 A selection differential is the phenotypic average difference of the
selected parent animals (Ps) from the population average (P'). Selection
intensity (i) is the selection differential expressed in terms of
phenotypic standard deviations (s); i.e. the ratio of (Ps-P') over s.
Selection is used in predicting genetic response due to selection because
it can be related to that percent of the population saved as parents
(see Table 1). If the top 100f goats available are used for selection,
their mean phenotypic superiority due to selection intensity is 1.75
standard deviations above the population mean. If the top 70are
selected then the selection intensity will be only 0.5 standard
deviations above average.
17 Table 1. Selection intensities (i) for different percentages of
individuals selected to be parents from a large population.
=========================================================================
Percent Saved i
-------------------------------------------------------------------------
1 2.67
5 2.06
10 1.75
30 1.16
50 0.80
70 0.50
90 0.19
100 0
18 The parameter of heritability is used to calculate selection response.
The square root of heritability is called accuracy (h). The formula for
selection response after using a certain superior male in a goat herd is
http://www.inform.umd.edu/EdRes/Topic/AgrEn...dd/goat/SELECTION_AND_BREEDING_SYSTEMS.html (4 of 10) [4/15/2004 12:34:25 PM]
SELECTION AND BREEDING SYSTEMS
one-half of the product of accuracy, times selection intensity, times the
additive genetic variance for the trait. The reason for ''one-half'' is,
of course, that only half of the sire's genetic superiority is passed on
to his progeny. Selection response is the genetic change due to
selection in one generation. Many times, interest lies in the genetic
change per year. To obtain this estimate, one must divide by the
generation interval (t), which is the average age of the parents when
their progeny are born. For example, if the heritability of yearling
weight is 0.49, then the accuracy is the square root or 0.7. The
generation interval of goats is two years. According to the above
formula, the answer in the case of 10election intensity (equal to 1.75
standard deviations) would be 1/2(0.7) (1.75)/2=0.31 standard deviations
selection response per year for strictly male selection only. If however,
700f all females selected is the selection intensity program, then the
selection progress becomes, according to the same formula,
1/2(0.7) (0.5)/2=0.09 standard deviations selection response per year
above herd average. If both programs are combined, the answers are
combined and the selection response for yearling weight becomes 0.4
standard deviations progress per year in average.
19 Selection for more than one trait based on independent culling levels
is accomplished by ranking candidates for both traits and requiring the
selected animals to meet a minimum standard in both traits. This process
is repeated each year with only minimum standards adjusted for progress
made by selection.
20 Figure 2 shows independent culling plotted for a large number of
animals for two traits. Goats in the upper righthand quadrant meet the
standards for both traits and are chosen to become parents of the next
generation. Two points in Figure 2 are goats A and B for the two traits.
Goat A is superior to the population for trait 1 but falls just below
the culling minimum level for trait 2; so this animal would be culled.
Goat B has a performance level just above the standards for both traits
and is kept as a parent. It can be seen that Goat A could probably be a
more appropriate parent than B. Their difference in performance for trait
2 is minimal while goat A is far superior to B for trait 1.
21 Compromises have to be made for goats which are superior for the one
trait but just below borderline for the other trait. This has led to the
procedure of Selection Index which allows ranking animals simultaneously
for two or more traits in one single index.
22 A selection index is calculated from the sum of trait means each
multiplied by appropriate factors which weigh their relative importance
genetically and economically. An index provides the opportunity for
multiple trait selection. For example, a goat breeder considers trait 1
two times as important as trait 2. Hence, trait 1 would have a weighting
factor double that of trait 2. In figure 2, goat A would now be selected
rather than goat B because of the index method. Selection index, under [ Pobierz całość w formacie PDF ]