"Frontmatter". In: Plant Genomics and Proteomics

M
ARKER
-A
SSISTED
S
ELECTION
Traditional plant breeding is based on the selection of plants based on visible
or measurable traits, the phenotype. Marker-assisted selection (MAS), on the
other hand, can be used provided the molecular markers are sufficiently
close to the gene under selection. A standard backcross breeding scheme is
shown in Figure 8.5, and that for a MAS is shown in Figure 8.6.
1 6 0
8. I
D E N T I F I C AT I O N A N D
M
A N I P U L AT I O N O F
C
O M P L E X
T
R A I T S

---

M
A R K E R
- A
S S I S T E D
S
E L E C T I O N
1 6 1
donor parent hh (DP)
(3n plants)
Self-pollinate
20 plants
Self-pollinate
200 plants
Self-pollinate
20 plants
Self-pollinate
420 plants
50 BCbF3plants
with hh seed
FINAL PRODUCT
Selfing generations
Initial cross
Repeated cycles of backcrossing
Pollinate
250 plants
25n plants
with hh
recurrent parent HH (RP)
(n plants)
After harvest, analyze 100
samples to identify 25 hh plants
Identify and pollinate
250 hh plants with
RP pollen
Select 200 best and
analyze to identify
50 hh individuals
that have desired
phenotype
Repeat process
until desired
number (b) of
backcrosses is
reached
(2n plants)
(20n plants)
(2n plants)
(5n plants)
(100n plants)
×
×
×
F1
F2
BC1F1
(2n plants)
BCbF1
F3
RP
×
×
(40n plants)
BCbF2
×
FIGURE 8.5.
Conventional backcross breeding scheme. (Adapted from Dreher et al.,
2000.)
Conventional breeding methods have produced impressive genetic
gains, but it is clear that this historical rate of improvement cannot be sus-
tained using these methods alone (Dreher et al., 2000; Dreher et al., 2003;
Morris et al., 2003). Therefore, the tools that have become available through

---

1 6 2
8. I
D E N T I F I C AT I O N A N D
M
A N I P U L AT I O N O F
C
O M P L E X
T
R A I T S
donor parent hh (DP)
(3n plants)
recurrent parent HH (RP)
(n plants)
×
×
×
F1
BC1F1
RP
(n plants)
(n plants)
Hh BC1F1
RP
BC2F1
BC3F1
Hh BC3F1
(11 plants)
(3n plants)
(n plants)
(n plants)
(11 plants)
BC3F2
hh BC3F2
(10 rows)
(53 plants)
hh BC3F2
13 BC3F3 ears
with hh ssed
Analyze13 samples
to confirm phenotype
Analyze 53 samples with 55
background markers to identify
13 plants with highest proportion
of the recurrent parent genome
Analyze 210 samples
with one marker
to identify 53 hh plants
Analyze 21 samples
with one marker
to identify 11 Hh plants
Repeat process
to obtaion BC3F1
plants
Analyze 21 samples
with one marker
to identify 11 Hh plants
Analyze one bulk sample from each
parental population using 165
background markers to select 55
markers for use in full MAS
Analyze 8 samples from each parental
population using 3 markers to identify
best marker
(13 plants)
(3n plants)
×
×
FINAL PRODUCT
Self-pollinate
13 plants
Self-pollinate
11 plants
MAS
MAS
MAS
MAS
Selfing generations
Initial cross
Repeated cycles of backcrossing
FIGURE 8.6.
MAS backcross breeding scheme. (Adapted from Dreher et al., 2000).
genomics will extend the repertoires available and complement conventional
breeding. MAS has the potential to increase the efficiency of a breeding
scheme by:
∑ Reducing the time required to develop a new variety 
∑ Lowering the size of populations needed, thereby eliminating large
and costly field evaluation 

---

C
ONVENTIONAL
B
REEDING
M
ETHODS
Conventional breeding methods involve three basic steps:
∑ The production of a population of plants incorporating the desirable
traits
∑ The evaluation and selection of the best within this population 
∑ The intercrossing of these superior individuals 
The new population is then passed through subsequent cycles of selection
and improvement. 
What happens for a trait that is only expressed in a seed, but does not
have a visible phenotype, for example, protein quality, in a conventional
breeding program? Because this is a seed trait the breeders cannot identify
individuals that have an improvement until the seed has been tested. There-
fore, they either have to wait until the end of the season to identify the desir-
able plants or carry through a much larger number of plants, many of which
will subsequently turn out to be useless. Because the trait will need bio-
chemical analysis, the superior individuals cannot be identified until some
time after the seeds are harvested. If the gene that is being introduced is
recessive it will not be expressed in the first generation, so the heterozygotes
cannot be identified. Therefore, the individuals also must be self-fertilized,
and a small number of progeny must be typed to identify the heterozygotes.
These cycles must be repeated to develop a new line with the introduced
gene in a genetic background as similar to the original commercial line as
possible.
I
MPLICATIONS OF
MAS.
Even if the plant’s phenotype under selection is a 
reliable indicator of the underlying genetic characteristic, the phenotypic
evaluation can be costly, time consuming, and affected by the growth 
environment. The use of molecular markers has the potential to provide 
a solution to these problems. Provided they are sufficiently closely linked 
to the desirable allele, the presence of such an allele can be determined
directly in seedlings by the use of these markers. This eliminates the costly
and time-consuming phenotypic evaluation. The molecular markers can 
also be used to distinguish between the homozygous and heterozygous
plants, thereby eliminating the need to self individuals to determine their
genetic makeup. An added benefit of MAS arises when a large number 
of molecular markers that cover the plant’s entire genome are used. 
These markers can then identify those individuals that contain the largest
contribution of genetic material from the recurrent recipient line, thereby
reducing substantially the number of generations required to introgress the
desired allele.
A comparison of conventional breeding methods and MAS has been
carried out (Dreher et al., 2000; Dreher et al., 2003; Morris et al., 2003). The
M
A R K E R
- A
S S I S T E D
S
E L E C T I O N
1 6 3

---

study identified a number of areas in which MAS would offer significant
advantages over conventional breeding methods. These advantages are:
∑ A reduction in the extent of phenotypic screening 
∑ The ability to identify the presence of multiple alleles related to a
single trait even if the alleles do not individually produce a detectable
influence on the expression of the traits 
∑ The ability to select multiple traits simultaneously 
∑ The screening of traits whose expression depends on the growth 
environment
∑ The early detection of superior lines, especially for seed-expressed
traits
∑ The ability to manipulate recessive genes and identify the 
heterozygotes
∑ A reduction in the number of backcrossing cycles 
Molecular markers are clearly a powerful tool for plant breeding. MAS
offers opportunities for reducing costs, saving time, and accomplishing
breeding goals unavailable through conventional methods. However, there
have been different levels of adoption of the MAS strategy between com-
mercial and publicly funded breeding efforts. Part of the explanation for the
relatively slow integration of MAS is the expense of establishing biotech-
nology research facilities, the difficulty of identifying useful markers, and
the identification of markers linked to traits of interest that are controlled by
a large number of minor genes. The continued development of new tech-
nology, including DNA chip technology and SNP detection, should over-
come some of these constraints. Especially important will be the ability to
study the expression of thousands of genes related to a trait of interest simul-
taneously and to dissect out that subset which is causative rather than con-
sequential. The identification of the genes that play dominant roles in plant
growth and development will also help narrow the focus in the search for
genes of importance in agronomic improvement.