Selection Methods for Genetics Help (page 2)
Artificial vs. Natural Selection
Artificial selection is operative when humans determine which individuals will be allowed to leave offspring (and/or the number of such offspring). Likewise, natural selection allows only those individuals to reproduce that possess traits adaptive to the environments in which they live. There are several methods by which artificial selection can be practiced.
If heritability of a trait is high, most of the phenotypic variability is due to genetic variation. Thus, a breeder should be able to make good progress by selecting from the masses those that excel phenotypically because the offspring-parent correlation should be high. This is called mass selection, but it is actually based on the individual's own performance record or phenotype. As the heritability of a trait declines, so does the prospect of making progress in improving the genetic quality of the selected line. In practice, selection is seldom made on the basis of one characteristic alone. Breeders usually desire to practice selection on several criteria simultaneously. However, the more traits selected for, the less selection "pressure" can be exerted on each trait. Selection should thus be limited to the two or three traits that the breeder considers to be the most important economically. It is probable that individuals scoring high in trait A will be mediocre or even poor in trait B (unless the two traits have a positive genetic correlation, i.e., some of the genes increasing trait A are also contributing positively to trait B). The breeder therefore must make compromises, selecting on a "total merit" basis some individuals that would probably not be saved for breeding if selection was being practiced on the basis of only a single trait.
The model used to illustrate the concept of genetic gain [Fig. 8-9(a)], wherein only individuals that score above a certain minimum value for a single trait would be saved for breeding, must now be modified to represent the more probable situation in which selection is based on the total merit of two or more traits [Fig. 8-9(b)].
In selecting breeding animals on a "total merit" basis, it is desirable to reduce the records of performance on the important traits to a single score called the selection index. The index number has no meaning by itself, but is valuable in comparing several individuals on a relative basis. The methods used in constructing an index may be quite diverse, but they usually take into consideration the heritability and the relative economic importance of each trait in addition to the genetic and phenotypic correlations between the traits.
An index (I) for three traits may have the general form
where a, b, and c are coefficients correcting for the relative heritability and the relative economic importance for traits A, B, and C, respectively, and where A', B', and C' are the numerical values of traits A, B, and C expressed in "standardized form." A standardized variable (X') is computed in a sample by the formula
where X is the record of performance made by an individual, is the average performance of the population, and s is the standard deviation of the trait. In comparing different traits, one is confronted by the fact that the mean and the variability of each trait is different and often the traits are not even expressed in the same units.
EXAMPLE 8.15 An index for poultry might use egg production (expressed in numbers of eggs per laying season), egg quality (expressed in terms of grades such as AA, A, B, etc.), and egg size (expressed in ounces per dozen).
EXAMPLE 8.16 An index for swine might consider backfat thickness (in inches), feed conversion (pounds of feed per pound of gain), and conformation score (expressing the appearance of the individual in terms of points from a standard grading system).
The standardized variable, however, is a pure number (i.e., independent of the units used) based on the mean and standard deviation. Therefore, any production record or score of a quantitative nature can be added to the score of any other such trait if they are expressed in standardized form.
Family and Pedigree Selection
When both broad and narrow heritabilities of a trait are low, environmental variance is high compared with genetic variance. Family selection (also referred to as kin selection) is most useful when heritabilities of traits are low and family members resemble one another only because of their genetic relationship. It is usually more practical to first reduce environmental variance before initiating selective breeding programs. Another way to minimize the effects of an inflated environmental variance is to save for breeding purposes all members of families that have the highest average performance, even though some members of such families have relatively poor phenotypes. In practice, it is not uncommon to jointly use more than one selection method; e.g., choosing only the top 50% of individuals in only the families with the highest averages.
Family selection is most beneficial when members of a family have a high average genetic relationship to one another but the observed resemblance is low. If inbreeding increases the average genetic relationship within a family more than the increases in phenotypic resemblance, the gain from giving at least some weight to family averages may become relatively large.
In this method, consideration is given to the merits of ancestors. Rarely should pedigree selection be given as much weight as the individual's own merit unless the selected traits have low inheritabilities and the merits of the parents and grandparents are much better known than those of the individual in question. It may be useful for characteristics that can only be seen in the opposite sex or for traits that will not be manifested until later in life, perhaps even after slaughter or harvest. The value of pedigree selection depends upon how closely related the ancestor is to the individual in the pedigree, upon how many ancestors' or colateral ancestors' records exist, upon how completely the merits of such ancestors are known, and upon the degree of heritability of the selected traits.
A progeny test is a method of estimating the breeding value of an animal by the performance or phenotype of its offspring. It has its greatest utility for those traits that (1) can be expressed only in one sex (e.g., estimating the genes for milk production possessed by a bull), (2) cannot be measured until after slaughter (e.g., carcass characteristics), or (3) have low heritabilities so that individual selection is apt to be highly inaccurate.
Progeny testing cannot be practiced until after the animal reaches sexual maturity. In order to progeny-test a male, he must be mated to several females. If the sex ratio is 1 : 1, then obviously every male in a flock or herd cannot be tested. Therefore, males that have been saved for a progeny test have already been selected by some other criteria earlier in life. The more progeny each male is allowed to produce the more accurate the estimate of his "transmitting ability" (breeding value), but, in so doing, fewer males can be progeny-tested. If more animals could be tested, the breeder would be able to save only the very best for widespread use in the herd or flock. Thus, a compromise must be made, in that the breeder fails to test as many animals as desired because of the increased accuracy that can be gained by allotting more females to each male under test.
The information from a progeny test can be used in the calculation of the "equal-parent index" (sometimes referred to as the "midparent index"). If the progeny receive a sample half of each of their parents' genotypes and the plus and minus effects of Mendelian errors and errors of appraisal tend to cancel each other in averages of the progeny and dams, then average of progeny = sire/2 + (average of dams)/2 or
Sire = 2(average of progeny)–(average of dams)
Practice problems for these concepts can be found at:
Add your own comment
- Kindergarten Sight Words List
- The Five Warning Signs of Asperger's Syndrome
- What Makes a School Effective?
- Child Development Theories
- Why is Play Important? Social and Emotional Development, Physical Development, Creative Development
- 10 Fun Activities for Children with Autism
- Test Problems: Seven Reasons Why Standardized Tests Are Not Working
- Bullying in Schools
- A Teacher's Guide to Differentiating Instruction
- Steps in the IEP Process