6/16/2023

Selection Selection
• Choosing which animals get to be parents. • Selection is an important tool for rearranging
genetic material to produce better fit individuals
• Choosing which male and female mate. for a particular purpose.

• So it is said as differential reproductive rate.

• Selection is of two types

✓ Natural
✓ Artificial.

1 2

Natural Selection
• In nature, the main force responsible for
selection is the principle “survival of the
fittest” in a particular environment.

• It is a very complicated process and many

factors determine the proportion of
individuals that reproduce.

3 4

Natural Selection Artificial Selection

• It is interesting to note that in the wild and even in • Artificial selection is that which is practiced by
domesticated animals to a certain extent, there is a man but here natural selection also play a part.
tendency towards an elimination of the defective or
detrimental genes that have arisen through • It is a powerful force to change the gene
mutations, or through natural selection. frequency in a certain population

• As a breeder and geneticist we have no interest in • It is also of two types:

natural selection. • Automatic and
• Deliberate (planned) selection

5 6

1
 6/16/2023

Consequences of artificial selection Hardy-Weinberg Law

• In large random mating populations gene and
genotypic frequencies will remain constant
generation after generation in the absence of
any external force.
• Equal fertility
• Equal sex ratio
• Equal opportunity for mating / fertilization
• No external force

7 8

Genetic Effects of Selection

• Selection does not create new genes. It is
practiced to increase the frequency of
desirable genes in a population and to
decrease the frequency of undesirable genes.
Suppose “A” is desirable gene and “a” is
undesirable gene.

9 10

Example Genetic Effects of Selection

• Suppose there are “A” and “a” genes present • These hybrid individuals (Aa) are allowed to
in a certain population and the frequency of mate with each other
each gene is 0.50. Let us have a cross between
the individuals having “A” and “a” genes. Aa X Aa
AA X aa • The progeny produced will be AA, 2Aa, and
Aa aa. Still the gene frequency will be same i.e.
0.50.
Frequency of “A” gene is 0.50 and of “a” gene is
also 0.50

11 12

2
 6/16/2023

Example Genetic Correlations among Traits

• Let us have selection for “A” gene and the • Whether or not the same genes responsible
individuals having undesirable gene (aa) are for qualitative inheritance of two or more
culled from the stock, traits

The frequency of “A” will be 0.67 and for “a” gene will • Some of the same genes responsible for
be 0.33. The increase in the frequency of “A” gene will quantitative inheritance affect two or more
increase the population of AA individuals. economic traits.
Hence individuals homozygous for desirable genes will
increase.

13 14

Causes of Genetic Correlation Genetic Correlations

• Plieotropy is the major • The genetic correlation between two traits may
cause of genetic correlation be very low, means that probably very few of the
same genes affect the two traits. e.g. type and
• Linkage is the second cause performance in beef cattle.
of genetic correlation. Closely
linked genes would tend to
stay together over several generations. • If selection on the basis of one trait not improve
the other, two traits are inherited independently.
• Genes farther apart on the same
chromosome would separate more • It means we have to do the selection for the both
Readily by crossing over in meiosis, and traits.
the relationship between traits would breakup.

15 16

Genetic Correlations BASIS OF SELECTION

• The genetic correlation ranges from –1 to + 1 through zero.
• Mass Selection
• The value +1 correlation Selection on the basis
means that two traits are
of (individuality) individuals own performance.
positively correlated if selection
for one trait improve the other
as a bonus. e.g. rate and efficiency
of weight gain in beef cattle.

• The value –1 correlation means that two traits are

genetically correlated in a negative manner. Selection
for the improvement of one trait will make decline in
the other trait. e.g. milk production and fat percentage

17 18

3
 6/16/2023

MASS SELECTION
• Family Selection • Mass selection is on the basis of individuality
✓ Selection on the basis of
Pedigree performance • Selection based on individuality means that an
✓ Selection on the basis animal is kept or rejected for breeding purposes
of collateral relatives on the basis of its own phenotype for a particular
performance trait, or traits.
(sibs, cousins, uncles and
aunts)
•  h2, most of the variation due to genetics
✓ Selection on the basis of • Selecting the individuals that excel phenotypically
progeny test can make good progress
• Selection on the basis of all available information
• As h2 , the prospects of progress decreases

19 20

Mass Selection Mass Selection

• Phenotype of an individual for a quantitative Mass Selection is practiced when
trait is not a true estimate of its genotype
(quantitative traits are affected both by ➢Heritability is high (> 0.30)
heredity and environment)
➢Observations can be made on both sexes
• And because no trait is 100 % heritable (birth and weaning weight)

➢Observation can be made early in life

• And environmental portion of phenotypic (birth and weaning weight)
superiority is not transmitted to the offspring

21 22

Mass Selection Mass

MassSelection
Selection
• Individuals are ranked in order of merit
• Selection for traits like milk production on the
basis of individuality is not successful as the • Some proportion of the population is culled.
observation can not be made early in life, & in
both sexes and h2 is not high (Av. 0.25)
• Biological observations usually follow normal
distribution (bell shaped curve)

23 24

4
 6/16/2023

Mass Selection Mass

MassSelection
Selection
• Estimation of mean and standard deviation of ➢Mean ± 1 SD covers 68% of the population
the population help to take selection
decisions. ➢Mean ± 1.5 SD covers 87.5% of population

• It is known that mean of the population (herd ➢Mean ± 2 SD covers 95% of the population
average etc.) divides the population into two
equal halves (50 % above and 50 % below) ➢Mean ± 3 SD covers 99% of the population

25 26

Mass Selection
➢If mean milk production of a herd is 2500 kg and
SD of 500 kg. 50% of the population will be saved
if all the cows/buffaloes yielding below 2500 kg
are culled.

➢The mean of the selected group will be certainly

higher.

➢This rise in the population mean of the next

generation (progeny) will approximately measure
the progress.

27 28

Mass
MassSelection
Selection Mass
MassSelection
Selection
➢ A population of 1000 beef animals has mean weaning • If the farmer decides to cull animals having less than
weight as 150 kg with a standard deviation of 10 kg. 130 kg weaning weight (Mean –2 SD), he has to
dispose of 25 animals. And the mean weaning weight
➢ Approximately 680 animals will have the weaning weight
ranging from 140 -160 kg (Mean ±1 SD) will rise to 150.5 kg.

➢ About 950 animals will fall within the range of 130 – 170 kg • In case of culling of animals having less than 140 kg
(Mean ± 2 SD).
(Mean –1 SD), he has to dispose of 160 animals and
➢ Whereas 25 animals will be having more than 170 kg of the mean weaning weight of the selected group will
weaning weight and only 5 animals will fall in category be 151.9 kg
having more than 180 kg weaning weight.

29 30

5
 6/16/2023

Av. Heritability estimates in dairy cattle

Genetic Evaluation & Breeding Value
Traits No. of Studies Average Range
FLMY 33 0.231 0.030 --- 0.580
• Actual Genetic worth of an individual for a
milk yield 10 0.250 0.030 --- 0.320 particular trait with reference to a base
Dry period 15 0.016 0.005 --- 0.490 population
AFC 30 0.260 0.010 --- 0.800
CI 21 0.174 - 0.010 --- 0.281
• Ultimate goal in animal breeding is to rank
SP 18 0.020 0.004 --- 0.357
animals according to their genetic merit
SPC 11 0.030 - 0.090 --- 0.050
Lifetime MY 10 0.119 0.087 --- 0.920
Herd life 6 0.067 0.037 --- 0.610 • And to use them efficiently for breeding
purposes

31 32

Genetic Evaluation & Breeding Value Genetic Evaluation & Breeding Value
• Hence genetic evaluation of animals is a key • Accuracy of breeding
values depends upon
issue to formulate any genetic improvement the precision of the
program performance recording

• Identification of genetically
superior animals is a master
key for any genetic improvement
program through selection
• For genetic evaluation of animals,
performance recording is a pre-requisite • Especially the males as males leave thousands of progeny

33 34

Genetic Evaluation & Breeding Value Genetic Evaluation & Breeding Value
• Actual breeding value • Estimated breeding values (EBVs) are calculated
for a particular trait from the phenotypic records
can not be known but
estimated breeding values • The phenotypic records are influenced by many
environmental factors besides the hereditary
(EBVs) are calculated factors

• For determination of precise breeding values,

these environmental influences are to be
minimized
• These are estimates of the true breeding values

35 36

6
 6/16/2023

Some environmental factors affecting

Genetic Evaluation & Breeding Value
milk yield
• Variations among the phenotypic records are • Measurements such as milk yield in dairy cattle are
affected by environmental influences such as
usually measured through a Statistical Term
➢Age of the cow at calving
“Variance”
➢Year and Season of calving
➢Preceding dry period
• Special statistical techniques have been ➢Frequency of milking
developed for partitioning of the variance into ➢Length of lactation
its components like Hereditary Variance and ➢Lactation number
Environmental variance • Corrections are made to eliminate these effects

37 38

Genetic Evaluation & Breeding Value Genetic Evaluation & Breeding Value
• For estimation of breeding values, knowledge of • EBV = h2 (Ps – P) (for single records)
heritability (h2 ), & repeatability is required
• EBV = N h2 (Ps – P) (For multiple records)
1 + (N-1) r
• Best Linear Unbiased Predictions (BLUP) where
procedure using the Individual Animal Model
(IAM) has become the world wide standard for N = Number of records
the prediction of breeding values of farm animals h2 = Heritability of the trait
r = Repeatability
• BLUP procedure minimizes all environmental P = Population average
influences Ps = Average performance of the selected
individual

39 40

MPPA & ERPA MPPA & ERPA

• In herds the females often differ in age or in where
number of lactations completed N = number of records
• It is difficult to rank them according to merit r = Repeatability
• Most Probable Producing Ability or Expected P = population/herd average
Real Producing Ability is derived to make valid Ps = average performance of the selected
comparisons individual
MPPA = P + Nr (Ps – P)
1 + (N-1) r

41 42

7
 6/16/2023

Rank the buffaloes in descending order. Milk yield in kg &

repeatability 0.4, h2 as 0.25 and the herd average 2500 kg Goal of selection
• Improve the frequency
of desirable alleles.

• Reduce frequency of
undesirable alleles.

• Eliminate deleterious genes.

43 44

What makes selection difficult? Methods ofselection

• Most traits of economic significance are quantitative 1. Single trait
(polygenic) - controlled by many genes. ▪ Generally not recommended
▪ Single trait selection can have unexpected and undesirable
• Phenotype ≠ Genotype We can’t see genes. consequences due to genetic correlations or environmental
responses.
• Sometimes, it’s difficult to separate genetics from
environmental influences.

• Some traits cannot be directly or easily selected for.

• Quantitative (population) genetics is about probabilities. E.g selection for only seed shape (round or wrinkled)

45 46

Continued… Multiple Trait Selection

1. Multiple trait 1. Tandem selection
▪ Involves more than one trait
2. Independen culling levels
at a single time.
▪ Usually recommended. 3. Selection index
▪ More successful in raising sheep
and/or goats in which improvement
of more than one trait is desirable.

47 48

8
 6/16/2023

Tandem Selection Tandem Selection

• Focus on one trait at a time until a satisfactory First Select for Udder size Then select for Litter size
level of performance is achieved; then move on
to another trait and so forth.

• Selection may result in changes (

positive or negative) to correlated traits.

– Milk yield vs. fat percentage (antagonistic)

– Staple length vs. fleece weight (favorable)

49 50

Advantage and Disadvantage of

Independent cullinglevels
Tandem Selection
Advantage • Select for two or more traits simultaneously.
• If the genetic correlation is positive the one
• Establish a minimum standard for each trait.
trait may also improve another.
Disadvantage
• Western Maryland Pasture-Based Meat Goat
• Possible to cull some genetic superior Performance Test Identifying genetically
individuals due to them falling a little below
superior bucks
the cut-off.
• It takes time and efforts, and it is the least
efficient method.

51 52

Effects of ICL’s
• Advantage:
• It is more practical than selection indexes
because it does not require the use of
complex equations.
• Decisions can be made early because
complete data on an individual is not needed.
• Top performing buck “Kiko” selected for excellent
carcass quality and parasitic resistance

53 54

9
 6/16/2023

Disadvantage: Selection Indices

• The primary problem with this method is that • Selection indexes have been considered and utilized for
many years in different forms
a producer may eliminate an individual that is
excellent in one trait due to poor performance • The use of an index requires the calculation of
in another. economic values or “weights” for each trait and then
putting the individual’s performance into an equation.

• This is generally not a large problem if the • The result of that equation is an index value for the
number of traits is kept low, generally two to animal and that index is then ranked to identify the
three traits. best animals for the combination of traits included in
the index.

55 56

Selection Indices Disadvantage of Selection Indices

• The problem with this system is that economic
values change continuously, and values for all
traits must be collected before selection can
occur.

• Because not all desired traits can be measured

at one time, this requires keeping animals
longer to collect all the necessary data.

57 58

10