GGEBiplot analysis of genotype × environment interaction in Agropyron intermedium

J. Bio. & Env. Sci. 2015
260 | Emami et al.
RESEARCH PAPER OPEN ACCESS
GGEBiplot analysis of genotype × environment interaction in
Agropyron intermedium
Elham Emami1
, Ezatollah Farshadfar1,2*
, Hooshmand Safari2
1
Department of Agronomy and Plant Breeding, Kermanshah Branch, Islamic Azad University,
Kermanshah, Iran
2
Campus of Agriculture and Natural Resources, Razi University, Kermanshah, Iran
Article published on April 21, 2015
Key words: Agropyron Intermedium, forage yield, stability, GGEbiplot
Abstract
In order to identify genotypes of Agropyron intermedium with high forage yield and stability an experiment was
carried out in the Research station of Kermanshah Iran.The 11 accessions were sown in a randomized complete
block design with three replications under rainfed and irrigated conditions during 2013-21-014 cropping deasons.
Combined analysis of variance indicated high significant differences for location, genotype and G × E interaction
(GEI) at 1% level of probability. Mean comparisons over environments introduced G4, G3 and G5 with maximum
forage yield over rainfed and irrigated conditions. Minimum forage yield was attributed to genotype G1.
GGEbiplot analysis exhibited that the first two principal components (PCA) resulted from GEI and genotype
effect justified 99.37% of total variance in the data set. The four environments under investigation fell into two
apparent groups: irrigated and rainfed. The presence of close associations among irrigated (E1 and E3) and
rainfed (E2 and E4) conditions suggests that the same information about the genotypes could be obtained from
fewer test environments, and hence the potential to reduce testing cost.The which-won-where pattern of
GGEbiplot introduced genotypes G3 and G4 as stable with high forage yield for rainfed condition, while G5 was
stable with high yield for irrigated condition. According to the comparison of the genotypes with the Ideal
genotype accessions G4, G3 and G9 were more favorable than all the other genotypes.
*Corresponding Author: Ezatollah Farshadfar  e_farshadfar@yahoo.com
Journal of Biodiversity and Environmental Sciences (JBES)
ISSN: 2220-6663 (Print) 2222-3045 (Online)
Vol. 6, No. 4, p. 260-267, 2015
http://www.innspub.net

261 | Emami et al.
Introduction
Agropyron with high forage yield and wide stability
in different climate especially drought and Salt
tolerance is one of the most important forage crops
(Sutka et al., 1995). Since, there is high variation
within and among different species of Agropyron, so
selection response for improving important traits is
high (Arghavani et al., 2010). Agropyron has been
applied in wide hybridization specially to transfer
alien genes into cultivated wheat (Farshadfar, 2012;
Xu, and Conner, 1994).
In crop breeding programs, genotypes are evaluated
in multienvironment trials (METs) for testing their
performance across environments and selecting the
best genotypes in specific environments. Genotype ×
environment (GE) interaction is an important issue
faced by plant breeders in crop breeding programs. A
significant GE interaction for a quantitative trait such
as grain yield can seriously limit progress in selection.
Variance due to GE interaction is an important
component of the variance of phenotypic means in
selection experiments (Hallauer et al., 2010).
GEI affects breeding progress because it complicates
the demonstration of superiority of any genotype
across environments and the selection of superior
genotypes (Magari and Kang, 1993; Ebdon and
Gauch, 2002). Another undesirable effect of GEI
includes low correlation between phenotypic and
genotypic values, thereby reducing progress from
selection. This leads to bias in the estimation of
heritability and in the prediction of genetic advance
(Comstock and Moll, 1963). Therefore, the magnitude
and nature of GEI determine the features of a
selection and testing program.
Yield data from regional performance trials, or more
generally, multi environment trails (MET), are
usually quite large, and it is difficult to understand
the general pattern of the data without some kind of
graphical presentation. The biplot technique provides
a powerful solution to this problem. A biplot that
displays the GGE of a MET data, referred to as a GGE
Biplot (graphical method), is an ideal tool for MET
data analysis (Yan, 2001; Yan and Hunt, 2001).
The GGE biplot analysis of these data showed that
ideal test environments could discriminate superior
performing from poor ones, and identify the target
areas. GGE biplot analysis was recently developed to
simultaneously use some of the functions of stability
methods. In phenotypic variation, E explains most of
the variation, and G and G × E are usually small (Yan,
2002). However, only G and G × E interaction are
relevant to cultivar evaluation, particularly when G ×
E interaction is determined as repeatable (Hammer
and Cooper, 1996). Hence, Yan et al. (2000)
deliberately put the two together and referred to the
combination as GGE. Following the proposal of
Gabriel (1971), the biplot technique was also used to
display the GGE of MET data, and is referred to as a
GGE biplot (Yan, 2001; Yan et al., 2000). The GGE
biplot is in fact a data visualization tool that
graphically displays G × E interaction in a two way
table (Yan et al., 2000). The GGE biplot is an effective
tool for the following applications: 1) Mega-
environment analysis (e.g.; “which won - where”
pattern), whereby specific genotypes can be
recommended for specific mega-environments (Yan
and Kang, 2003). 2) Genotype evaluation (mean
performance and stability), and 3) Environmental
evaluation (to discriminate among genotypes in target
environments). GGE biplot analysis is increasingly
being used in G × E interaction studies in plant
breeding research (Butron et al., 2004; Dehghani et
al., 2006; Kaya et al., 2006; Samonte et al., 2005;
Yan and Tinker, 2005).
The objectives of this study were (i) to interpret G
main effect and GE interaction obtained by combined
analysis of yield performances of 11 Agropyron
intermedium over 4 environments (ii) application of
the GGE biplot technique to identify stable and high
yielding genotypes.
Materials and methods
In order to evaluate phenotypic stability of forage

262 | Emami et al.
yield 11 accessions of Agropyron intermedium were
prepared from gene bank of Research Institute of
Forests and Rangelands, Tehran, Iran (Table 1).
The experiment was carried out in the Research
station of Kermanshah Iran (47° 20´ N latitude, 34°
20´ E longitude and 1351.6 m altitude). Climate in the
region is classified as semiarid with mean annual
rainfall of 378 mm. Minimum and maximum
temperature at the research station were -27°C and
44°C, respectively.
The genotypes were sown in a randomized complete
block design with three replications under rainfed
and irrigated conditions during 2013-21-014 cropping
deasons. Each replication consisted of 11 genotypes
with 2 m length and 1 m wide and the distance
between two plots was 75 cm. Single seeds were
planted in 4 rows with 25 cm distance. Each plot
consisted of 3 rows with 1 m in length and 20-cm row
spacing. Data on forage yield were taken from all rows
of each plot. At harvest forage yield was determined
for each genotype at each test environments.
Statistical analysis
Analysis of variance on grain yield was conducted by
Genstat software to determine the effect of
environment (E), genotype (G) and GE interaction.
Coefficients between pairs of locations were
computed via SAS 9.2 software. The first two
components resulted from principal components
were used to obtain a biplot by GGE biplot software
(Yan, 2001). The basic model for a GGE Biplot is:
(1)
Where ijY = the mean yield of genotype i(=1,2,…,n) in
environment j(=1,2,…m),  = the grand mean, j =
the main effect of environment j, ( j  ) being the
mean yield of environment j, l = the singular value
(SV) of lth principal component (PC), the square of
which is the sum of squares explained by
PCl=(l=1,2,…,k with k≤ min (m,n) and k=2 for a two-
dimensional biplot), il = the eigenvector of genotype
i for PCl, lj = the eigenvector of environment j for
PCl, ij = the residual associated with genotype i in
environment j. To generate a biplot that can be used
in visual analysis of MET data, the SVs have to be
partitioned into the genotype and environment
eigenvector so that the model (1) can be written in the
form of 

k
i
ijljiljij egY
1
 where gil and
elj are called PCl scores for genotype i and
environment j, respectively. In a biplot, genotype i is
displayed as a point defined by all gil values, and
environment j is displayed as a point defined by all elj
values (l=1 and 2 for a twodimensional biplot) (Yan
and Kang, 2003).
Results and discussion
Combined analysis of variance and mean
comparisons
Combined analysis of variance indicated high
significant differences for location, genotype and G ×
E interaction (GEI) at 1% level of probability (Table
2). But maximum contribution of variance was
observed for location (70.34%). In the
multienvironment experiment the contribution of
environment (location and year) is more than G and
GEI (Farshadfar et al., 2012). Farshadfar (2012)
reported that in the Agropyron species different
water potential in the irrigated and rainfed conditions
accounted for maximum contribution of location.
Significant difference between the genotypes
indicating that selection for forage yield is desirable
for introduction of high yielding accessions.
Significant GEI with 6.62% of contribution in the
total sum of squares (SS) exhibiting that we can
proceed and calculate phenotypic stability in the
genotypes under investigation. Least contribution of
year effect in the total SS (0.02%) revealed that the
effect of year on the forage yield is low.

263 | Emami et al.
Mean comparisons over environments introduced G4
(5284g), G3 (5079g) and G5 (5043g) with maximum
forage yield over rainfed and irrigated conditions.
Minimum forage yield was attributed to genotype one
(G1=3708g).
Table 1. Numbers and codes of the agropyron
intermedium accessions investigated.
Accessions codes Numbers
890-2 1
890-4 2
890-5 3
890-6 4
890-7 5
890-9 6
890-10 7
890-11 8
89013 9
890-14 10
890-15 11
GGEbiplot analysis of phenotypic stability
The GGE biplot graphically displays G plus GE of a
MET in a way that facilitates visual cultivar evaluation
and mega environment identification (Yan et al.,
2000). Only two PC (PC1 and PC2) are retained in the
model because such a model tends to be the best
model for extracting patterns and rejecting noise from
the data. In addition, PC1 and PC2 can be readily
displayed in a twodimensional biplot so that the
interaction between each genotype and each
environment can be visualized (Yan and Hunt , 2001).
GGEbiplot was employed to identify stable genotypes.
The first two principal components (PCA) resulted
from GEI and genotype effect justified 99.37% of total
variance in the data set. PCA1 and PCA2 explained
81.75% and 17.62% of variability respectively.
Table 2. Combined analysis of variance.
S.O.V D.F SS % of TSS MS
Location (L) 1 202898725 70.34 202898725**
Error 1 4 2305838 0.80 576460
Genotype (G) 10 31884997 11.05 3188500**
G×L 10 19094158 6.62 1909416**
Error 2 40 26558028 9.21 663951
Year (Y) 1 66286 0.02 66286 ns
Y×L 1 55884 0.02 55884 ns
Y×G 10 114008 0.04 11401 ns
Y×G×L 10 381308 0.13 38131 ns
Error 3 44 5091701 1.77 115720
**: Significant at 1% probability level; ns: non-significant.
Relationships among test environments
GGE biplot, which was based on environment focused
scaling, was portrayed to estimate the pattern of
environments (Fig. 1). The vector view of the GGE
biplot (Fig. 1) provides a summary of the
interrelationships among the environments. The lines
that connect the test environments to the biplot origin
are called environment vectors. The cosine of the
angle between the vectors of two environments
approximates the correlation between them. For
example, E1 and E3 (irrigated conditions) and E2 and
E4 (rainfed conditions) were positively correlated (an
acute angle). But irrigated (E1 and E3) and rainfed
(E2 and E4) conditions were not correlated (a right
angle) indicating the environmental diversity and
independent in genotype rankings.
The distance between two environments measures
their dissimilarity in discriminating the genotypes.
Thus, the four environments fell into two apparent
groups: irrigated and rainfed. The presence of close
associations among irrigated (E1 and E3) and rainfed
(E2 and E4) conditions suggests that the same
information about the genotypes could be obtained

264 | Emami et al.
from fewer test environments, and hence the
potential to reduce testing cost. If two test
environmens are closely correlated consistently
across years, one of them can be dropped without loss
of much information about the genotypes.
Fig. 1. GGE biplot based on relationships among test environments.
Fig. 2. Polygon views of the GGE biplot based on symmetrical scaling for the which-won-where pattern of
genotypes and environments.
Which Won Where Pattern
One of the most attractive features of a GGE biplot is
its ability to show the which-won-where pattern of a
genotype by environment dataset (Fig. 2). Many
researchers find this use of a biplot intriguing, as it
graphically addresses important concepts such as
crossover GE, mega environment differentiation,
specific adaptation, etc (Yan and Tinker, 2006). The
polygon is formed by connecting the markers of the
genotypes that are further away from the biplot origin
such that all other genotypes are contained in the
polygon. Genotypes located on the vertices of the
polygon performed either the best or the poorest in
one or more environments since they had the longest
distance from the origin of biplot. The perpendicular
lines are equality lines between adjacent genotypes on
the polygon, which facilitate visual comparison of
them. These lines divide the biplot into 4 sectors, and
the environments fall into 2 of them (Fig. 2). An
interesting feature of this view of a GGE biplot is that
the vertex genotype(s) for each sector has higher
(some times the highest) yield than the others in all
environments that fall in the sector (Yan, 2002).
Thus, E2 and E4, fell into sector 1 and the vertex

265 | Emami et al.
genotypes for this sector were G3 and G4, suggesting
that they are stable with high forage yield and
adapted with rainfed conditions. Similarly, E1 and E3,
fell into sector 2 and the vertex genotype for this
sector was G5, suggesting that the higher-yielding
genotype with adaptability for irrigated condition was
G5. Sector 3 included G1, G2, G6, G7, G8 and G10
which showed no specific adaptation to any
environment indicating their low yield in both
conditions. Sector 4 with rainfed environment
contained no genotypes.
Fig. 3. Average environment coordination (AEC) views of the GGE-biplot based on environment-focused scaling
for the means performance and stability of genotypes.
Performance and stability of genotypes were
visualized graphically through the GGE biplot (Fig. 3).
In Figure 3 X-axis is an indicator of forage mean
yield, while Y-axis exhibits stability of genotypes.
Therefore it is possible to identify simultaneously
genotypes with high yield with stability. It is to be
mentioned that in GGEbiplot the effect of G + GEI are
considered simultaneously and not separated from
each other, that, s why the line has an ascending
order. If the contribution of G and GEI in the variance
is equal, then the horizontal and central line will be
parallel to X-axis, but here the contribution of GEI is
less than G and this is the cause of line ascending
order, to avoid confounding effect of G and GEI, and
proportion of variance of PC1(which is usually the
effect of genotype) go for GGI and therefore the effect
G and GEI separate from each other (Yan et al.,
2009a). Yan et al. (2009b) reported that the ideal
genotype has high PC1(high yield) and low PC2 (high
stability). However yield and stability of accessions
can be evaluated by average environment
coordination (AEC) method (Yan, 2001; 2002). In
Fig. 3 the line with single arrow head is the AEC
(average environment coordinate) abscissa. AEC
abscissa passes through the biplot origin and marker
for average environment and points towards higher
mean values. The average environment has average
PC1 and PC2 scores across environments (Yan, 2001).
The perpendicular lines to the AEC passing through
the biplot origin are referred to as AEC ordinate. The
greater the absolute length of the projection of a
genotype indicates more instability. Furthermore, the
average yield of genotypes is approximated by the
projections of their markers to the AEC abscissa (Yan
and Kang, 2003). According to Fig. 3, genotypes with
above average means were from G9, G3 and G4, while
genotypes below-average means were from G8 to G1.
However, the length of the average environment
vector was sufficient to select genotypes based on
yield mean performances. Genotypes with above-
average means (G9, G3, G4 and G5) could be selected,
whereas the rest were discarded. A longer projection
to the AEC ordinate, regardless of the direction,

266 | Emami et al.
represents a greater tendency of the GE interaction of
a genotype, which means it is more variable and less
stable across environments or vice versa. For
instance, genotype G3 was more stable as well as high
yielding followed by G4. Conversely, G5 was instable,
but high yielding.
Fig. 4. GGE biplot based on genotype-focused scaling for comparison the genotypes with the ideal genotype.
Comparison of the Genotypes with the Ideal
Genotype
An ideal genotype have the highest mean
performance and be absolutely stable (i.e., perform
the best in all environments). Such an ideal genotype
is defined by having the greatest vector length of the
high-yielding genotypes and with zero GE, as
represented by the small circle with an arrow pointing
to it (Yan, 2001). Although such an ideal genotype
may not exist in reality, it can be used as a reference
for genotype evaluation. A genotype is more desirable
if it is located closer to the ideal genotype. Thus, using
the ideal genotype as the center, concentric circles
were drawn to help visualize the distance between
each genotype and the ideal genotype (Fig. 4). In Fig.
4 the genotypes are ranked relative to the ideal
genotype. A genotype is more favorable if it is closer
to the ideal genotype. Accordingly, genotypes G4, G3
and G9 were more favorable than all the other
genotypes. The other genotypes were unfavorable
because they were far away from the ideal genotype.
References
Arghavani A, Asghari A, Shokrpour M,
Mohammaddost C. 2010. Genetic Diversity in
Ecotypes of two Agropyron Species using RAPD
Markers. Research Journal of Environmental
Sciences 4, 50-56.
Butrón A, Velasco P, Ordás A, Malvar RA.
2004. Yield evaluation of maize cultivars across
environments with different levels of pink stem borer
infestation. Crop Science 44, 741-747.
Comstock RE, Moll RH. 1963. Genotype ×
environment interactions. Symposium on Statistical
Genetics and Plant Breeding. National Academy
Science and National Research Council. Washington
D.C., USA, 164-196.
Dehghani H, Ebadi A, Yousefi A. 2006. Biplot
analysis of genotype by environment interaction for
barley yield in Iran. Agronomy Journal 98, 388-393.
Ebdon JS, Gauch HG. 2002. Additive main effect
and multiplicative interaction analysis of national
turfgrass performance trials: I. Interpretation of
Genotype × Environment interaction. Crop Science
42, 489-496.
Hallauer RA, Carena MJ, Miranda JB. 2010.
Quantitative Genetics in Maize Breeding, Springer
New York. R.W.

267 | Emami et al.
Hammer GL, Cooper M. 1996. Plant Adaptation
and Crop Improvement. CAB International. 608 p.
Kaya Y, Akcura M, Taner S. 2006. GGE-biplot
analysis of multi-environment yield trials in bread
wheat. Turkish Journal of Agriculture 30, 325-337.
Farshadfar M. 2012. Genetic variability and
Karyotype analysis of some Agropyron species.
Annals of Biological Research 3(3), 1515-1523.
Magari R, Kang MS. 1993. Genotype selection via a
new yield stability statistic in maize yield trials.
Euphytica 70, 105-111.
Samonte SOPB, Wilson LT, McClung AM,
Medley JC. 2005. Targeting cultivars onto rice
growing environments using AMMI and SREG
GGEbiplot analysis. Crop Science 45(6), 2414-2424.
Sutka J, Farshadfar E, Kosegi B, Friebe B, Gill
BS. 1995. Cereal Research Comunications 4, 351-357.
Xu J, Conner RL. 1994. Intravarietal variation in
satellites and C-banded chromosomes of Agropyron
intermedium ssp. trichophorum cv. Greenleaf.
Genome 37(2), 305-310.
Yan W. 2001. GGEBiplot - A Windows application
for graphical analysis of multi environment trial data
and other types of two-way data. Agronomy Journal
93, 1111–1118.
Yan W. 2002. Singular-value partitioning in biplot
analysis of multi-environment trial data. Agronomy
Journal 94, 990-996.
Yan W, Hunt LA. 2002. Biplot analysis of diallel
data. Crop Science 42(1), 21-30.
Yan W, Hunt LA, Sheng Q, Szlavnics Z. 2000.
Cultivar evaluation and mega-environment
investigation based on the GGE biplot. Crop Science
40, 597-605.
Yan W, Kang MS. 2003. GGE biplot analysis: a
graphical tool for breeders, geneticists and
agronomist. CRC Press, Boca Raton, FL. 271 p.
Yan W, Kang MS, Ma B, Wood S, Cornelius
PL. 2009a. GGE biplot vs. AMMI analysis of
genotype-by-environment data. Crop Science 47,
643-655.
Yan W, Tinker NA. 2005. An integrated system of
biplot analysis for displaying, interpreting, and
exploring genotype by environment interactions.
Crop Science 45, 1004-1016.
Yan W, Tinker NA. 2006. Biplot analysis of multi-
environment trial data: Principles and applications.
Canadian Journal of Plant Science 86, 623-645.
Yang RC, Crossa J, Cornelius PL, Burgueno J.
2009. Biplot Analysis of Genotype × Environment
Interaction: proceed with Caution. Crop Science 49,
1564-1576.

GGEBiplot analysis of genotype × environment interaction in Agropyron intermedium

More Related Content

What's hot (20)

Viewers also liked (19)

Similar to GGEBiplot analysis of genotype × environment interaction in Agropyron intermedium (20)

More from Innspub Net (20)

Recently uploaded (20)

GGEBiplot analysis of genotype × environment interaction in Agropyron intermedium