Molecular Plant Breeding 2015, Vol.6, No.21, 1
-
17
2
uniform. When both the seed purity and SSR-HLR
exceed the threshold values, the variety is deemed to
be uniform (Wang et al., 2009a, 2009b; Liu et al.,
2013; Wang et al., 2014a, 2014b).
Each wheat variety is a self-pollinating population
derived from the offspring of crossing two parents.
For each selfing event in an individual wheat plant, a
heterozygous locus will segregate with a 1:2:1 ratio in
the offspring, resulting in a non-homozygous locus in
the offspring population. To differentiate it from a
heterozygous locus of an individual plant, we define a
non-homozygous locus as a locus with maternal and
paternal homozygous genotypes as well as the parental
heterozygote (aa, bb, and ab) among individuals of a
line or a given variety. To differentiate homozygous
locus of a plant, we define a homozygous locus of a
line or variety as a locus with the same homozygous
genotype among all individuals of a given variety.
Non-homozygous loci are inherent in wheat varieties.
The proportion of non-homozygous loci in the
population of each later generation will be reduced
because successive generations are derived from a
selfing plant (individual) during breeding, but
concurrently the HLR will increase. Wang et al.
(2009b, 2014a) document the SSR-HLR of F
4
, F
5
and
F
6
wheat lines as 84.7-94.8%, 96.1-99.4% and ≥98%,
respectively, proving that most wheat varieties have
non-homozygous loci; some morphological traits of F
4
lines still segregate and inter-plant phenotypic differences
are obvious; morphological trait segregation in most
wheat lines ends at the F
5
or F
6
generations and thus
these lines show morphological and agronomic
uniformity even if they have non-homozygous loci.
Breeders will choose the lines that perform best in
terms of agronomic traits and raise the successive
generations from those lines. At this point, the
SSR-HLR in the population will become fixed. For
example, if the SSR-HLR of an F
5
plant is 95% of the
genome, the SSR-HLR of the line derived from that
plant will be 95%, and the SSR-HLR of the variety or
cultivar derived from the line will be 95%. If the
SSR-HLR of an F
4
plant is 90% of the genome, the
SSR-HLR of the line derived from that plant will be
90%, and the SSR-HLR of the variety or cultivar
derived from the line will be 90% and inter-plant
phenotypic differences are obvious.
However, inter-plant genotypic differences are caused
not only by non-homozygous loci, but also by
contaminant individuals (Wang et al., 2009b, 2014b).
The contaminant individuals are from other varieties
or hybrids with other varieties, and their genotypes at
some loci are different from those of the variety being
tested and are easily identified when DNA markers
are used to genotype a number of loci. Therefore, it is
essential to distinguish the genotypes of contaminant
individual from non-homozygous SSR loci in
evaluation of SSR-HLR (Wang et al., 2014a).
The key to evaluating seed purity is accurate
identification of contaminant individuals. To accurately
identify contaminant individuals, it is also essential to
distinguish contaminant individual genotypes from
non-homozygous SSR loci. In previously published
studies on the assessment of crop seed purity with
molecular markers, most authors attribute all inter-plant
genotypic differences to contaminant individuals (see
the references of Wang et al., 2014b). It is likely that
the estimated seed purity is lower than the actual
purity. To resolve the problem, we have proposed the
method for discriminating two phenomena (Wang et
al., 2014b). The methods for estimating seed purity
and SSR-HLR of wheat varieties, based on
discrimination of contaminant individual genotypes
from non-homozygous SSR loci, are described
elsewhere (Wang et al., 2014a, 2014b). This paper
outlines the technical procedures and criteria for
uniformity assessment based on both seed purity and
SSR-HLR.
1 Results
Impact of non-homozygous SSR loci on the uniformity
of wheat varieties
A variety is not uniform if contaminants and
non-homozygous SSR loci exceed a certain proportion.
Of 633 national winter wheat variety regional trials in
China, about 10% of the varieties had ≥ 10%
non-homozygous SSR loci. For example, 73 of 91
SSR loci of the variety A2009-75 were homozygous
and the other 18 SSR loci were non-homozygous,
resulting in a low SSR-HLR (80.2%). Of 18
non-homozygous loci, 11 loci had two homozygous
genotypes (aa and bb) and the heterozygote (ab)
among 20 individuals analyzed, and seven loci had
only two homozygous genotypes due to small sample
size. Number of individuals with different genotypes
at eight of the 18 loci in 20 individuals is presented in
