Introduction
Soybean [Glycine
max (L.) Merr.] is the “Miracle legume” owing to
its extraordinary oil, protein, vitamin and mineral contents with several
health benefits (Vishwanath et al.
2021). It provides several vital raw materials for many industries thereby
contributing significantly to the Indian economy. Agriculturally, its
cultivation substantially enriches soil fertility by atmospheric nitrogen
fixation (Jain et al. 2018). The genetic yield potential of soybean can
be realized in the Indian sub-continent through (i) characterization
of various germplasm collections maintained at different regions (ii)
identification of trait-specific adaptable accessions and (iii) utilization of
such identified genotypes in the targeted pre-breeding programs. The
geographical situation makes a shift in the phenological pattern and yield in
soybean (Bisen et al. 2015). Further, a narrow
genetic base and stress susceptibility reduce yield in soybean (El-Harty et
al. 2018) and therefore necessitate the identification of region-specific
donors for yield improvement.
Pre-breeding by utilizing such genotypes gives an
exceptional opportunity to broaden the genetic base and thereon evolution of
potential segregants becomes possible (Sharma et al. 2013). A core set
of germplasm represents the total variability of that respective collection and
it facilitates precise pre-breeding activities (Upadhyaya et al. 2009). Muthamizhan et al. (2016) developed a core set
representing the soybean germplasm pool of Tamil Nadu Agricultural University
whose variability pattern, trait dependability, and adaptability are to be
assessed. The utility of quantitative traits in core set characterization is
well documented (Upadhyaya et al. 2009). The qualitative traits are
helpful in the identification of morphological markers which eases the
selection process in different plant breeding cycles and provide reliable
opportunities for varietal characterization. The Principal Component Analysis
(PCA) and association analysis are helpful in tagging desirable genotypes and traits
with their contribution to the total divergence (Mohan et al. 2019).
Yadav (2016) established the utility of the morphological descriptors (NDUS –
Novelty Distinctness Uniformity and Stability) for soybean cultivar
identification. The usefulness of R language and UPGMA in germplasm
characterization were well reported (Epskamp et al.
2019).
The production of soybean in India has been drastically
reduced over the years due to poorly adapted genotypes and rainfed cultivation
(Rao and Chaitanya 2020) where Tamil Nadu is not an exception. The TNAU soybean
germplasm is constituted over a period of time (1975-2018) through inclusion of
various trait specific exotic and indigenous genotypes with a view to develop
location specific germplasm. A TNAU soybean core set was developed first of its
kind in Tamil Nadu soybean improvement programs, whose validation is the need
of the hour to improve soybean productivity in Tamil Nadu, India. Therefore,
the present study aimed to evaluate the adaptability of a TNAU soybean core
set, identification of desirable trait specific genotypes, and their validation
through targeted hybridization programs.
Materials and
Methods
Experimental Materials
The experimental material comprised of a soybean core
collection of 43 accessions and seven checks developed from a global germplasm
collection maintained at Dr. Ramaiah gene Bank, Tamil Nadu Agricultural
University (TNAU), Coimbatore (Muthamizhan et al.
2016). The core set includes thirty National and twenty international genotypes
whose geographical origin is depicted in Fig. 1.
Experimental Season
and Location
The preliminary core set development experiment was
conducted at the Department of Pulses, Centre for Plant Breeding and Genetics,
TNAU, Coimbatore, during Kharif season 2016. The adaptability of the
core set was confirmed through second and third-year confirmative experiments
during Rabi 2016 and summer
2017, respectively.
Experimental
Design and Data Documentation
The experiments were conducted in a Randomized Block
Design (RBD) with three replications. The spatial pattern was 30 cm × 10 cm.
The genotypes were grown in four-meter rows. All the recommended packages of
practice were followed to raise healthy crops as set forth in Crop Production
Guide for Tamil Nadu (TNAU 2016). Both quantitative and qualitative traits were
utilized to assess the potency of the soybean core set.
Quantitative Trait-based
Experiments
For quantitative trait based PCA, in each replication ten
plants were randomly selected per genotype and ten following quantitative
characters plant height, days to fifty percent flowering, days to maturity,
leaf area, number of primary branches, number of clusters per plant, number of
pods per plant, number of seeds per pod, hundred seed weight, and single plant
yield were recorded. Days to fifty percent flowering and maturity were recorded
at appropriate stages. Leaf area was measured at the third fully opened leaf
from meristem. The other traits were measured at harvest.
Qualitative Trait-based
Experiments
For qualitative trait-based cluster analysis, in each
replication ten plants were randomly selected per genotype and observations on
twenty-one qualitative characters (Table 1) were documented as per the soybean descriptor of Bioversity International at appropriate growth stages
(Ramteke and Murlidharan 2012) and used for
clustering.
Validation of
the Soybean Core Set
To test verify the potential of the TNAU soybean core
set, a set of contrast and closely related parents were selected for effecting
hybridizations during Kharif, 2017.
The parents were selected based on the results of quantitative trait based PCA,
qualitative trait-based similarity index, and yield performance. The genotypes
that are scattered far apart in PCA biplot were tagged as diverse and the
accessions located near the origin were considered as closely related parents.
The genotypes with higher and lower similarity indices were treated as closely
related and vice versa. Accordingly,
the genotypes JS(SH)99-02, IC16009, TNAU20048 and JS9305 were selected as more
diverse (group I) while the accessions NRC77, UGM75, EC30198 and Co (Soy)3 were
earmarked as closely related (group II). Owing to yield potential, the
genotypes Co (Soy)3 and IC16009 were utilized as females in the group I and II
respectively while other genotypes were used as male parents. In each group,
three crosses were made. The progenies of crosses were handled separately. The
F1 seeds (20–30 seeds per cross) were sown during Rabi, 2017. The true F1s
were identified based on the morphological characters used in the confirmative
experiments and forwarded to F2 generation (200–250 seeds per cross)
during Kharif, 2018. The performances
of F2 segregants were quantified using the ten quantitative characters and
compared.
Statistical Analysis
The quantitative
data documented during the mentioned three seasons were pooled using Pbtools software (http://bbi.irri.org/). The PCA (Jackson
and Edward 1991) for the pooled data was performed using ‘factoextra’
and ‘FactomineR’ packages of R studio version 1.0.136
(Lę et al. 2008). The correlation and path analysis were done as Table 1: The
list of qualitative characters used for characterization of the TNAU soybean
core set
|
S. No |
Qualitative traits |
S. No |
Qualitative traits |
|
1 |
Leaflet shape |
10 |
Plant height |
|
2 |
Petiole presence |
11 |
Hilum colour |
|
3 |
Pubescence density |
12 |
Hypocotyl anthocyanin
pigmentation |
|
4 |
Days to 50% flowering |
13 |
Pod pubescence colour |
|
5 |
Flower colour |
14 |
Seed cotyledon colour |
|
6 |
Growth type |
15 |
Seed size |
|
7 |
Pod colour |
16 |
Seed lusture |
|
8 |
Pod pubescence |
17 |
Plant growth habit |
|
9 |
Days to maturity |
18 |
Seed coat colour |
|
|
Biochemical based tests |
|
|
|
19 |
Seed coat peroxidase activity |
|
|
|
20 |
Sodium hydroxide test |
|
|
|
21 |
Potassium hydroxide test |
|
|
%20184-192_files/image002.jpg)
Fig. 1: Geographical origin of the TNAU soybean core set
suggested by Pearson (1901) and Dewey and Lu (1959),
respectively. The visualization of correlation and path analysis were obtained
from ‘corrplot’ and ‘semPlot’
packages of R studio version 1.0.136 (Wei et al. 2017; Epskamp et al.
2019). The genetic associations between genotypes based on qualitative
data were estimated by Jaccard’s similarity coefficient. The similarity matrix
was used to group the genotypes by Sequential Agglomerative Hierarchical
Non-overlapping (SHAN) clustering technique utilizing the Unweighted Pair Group
Method with Arithmetic Averages (UPGMA) method. The statistical analysis for
qualitative data was carried out using NTSYSpc
software (version 2.2).
Results
Both quantitative and qualitative traits were utilized
in the current investigation to understand the breeding value of the soybean
core set and the results were validated through a hybridization experiment.
Quantitative Trait-based
Experiments
Principal Component Analysis (PCA)
The PCA was performed to know the genetic relatedness of
the genotypes, interdependence of various traits and importance of traits with
respect to total variation. The quantitative trait based PCA divided the total
variation into ten Principal Components (PCs). The first three PCs contributed
47.63, 15.66 and 10.43 percent to the total variation, respectively and
therefore considered as major principal components (Table 2). The genotypes
were scattered along the biplot based on the first two PCs (Fig. 2). The
genotypes viz., JS9305, JS(SH)99-02,
IC16009 and TNAU20048 were located far apart, while the genotypes NRC77, UGM75,
EC30198 and MAUS61 were placed closer to the origin. The interrelationship and
contribution of quantitative characters to the total variation are represented
in Fig. 3. The characters single plant yield, number of pods per plant and
plant height were away from the origin that %20184-192_files/image004.png)
Fig. 2: Genetic divergence of TNAU
soybean core set in biplot with cos2 loadings
*circled
in red color are diverse parents and in green are closely related parents used
for hybridization
contributed
the maximum to the divergence. On the Table 2: Contribution of
ten principal components to the total divergence with Eigen values
|
Principal
components (PC) |
Eigen value |
Percentage of
variance |
Cumulative
percentage of variance |
|
PC1 |
4.76 |
47.63 |
47.63 |
|
PC2 |
1.57 |
15.66 |
63.29 |
|
PC3 |
1.04 |
10.43 |
73.72 |
|
PC4 |
0.83 |
8.34 |
82.06 |
|
PC5 |
0.74 |
7.44 |
89.50 |
|
PC6 |
0.42 |
4.23 |
93.73 |
|
PC7 |
0.29 |
2.95 |
96.68 |
|
PC8 |
0.20 |
1.97 |
98.65 |
|
PC9 |
0.10 |
1.02 |
99.67 |
|
PC10 |
0.03 |
0.33 |
100.00 |
%20184-192_files/image006.png)
Fig. 3: Variables plot with contribution of quantitative
characters to the total divergence
contrary,
leaf area and number of seeds per plant were closer to the origin and contributed
the minimum. Further, the contributions of various quantitative characters to
various PC’s were analyzed to understand their importance and are depicted in a
rotated component matrix (Fig. 4). Single plant yield (0.92) and number of pods
per plant (0.92) contributed the maximum for the rotated PC1. Similarly, days
to maturity (0.97) and number of seeds per pod (0.96) contributed the maximum
for the rotated PC2 and PC3, respectively.
Genetic Association Studies
Even though,
PCA hints the relationship between the quantitative characters, the magnitude
of the relationship can be arrived from correlation and path analysis. The
correlation and path analyses were performed to decipher the influence of
various quantitative traits on single plant yield and the results are depicted
in Fig. 5 and 6. The association analysis revealed that the number of pods per
plant (r = 0.88) contributed the maximum for single plant yield. It was followed
by number of clusters per plant (r =
0.81), plant
height (r = 0.70) and number of primary branches (r = 0.64). The number of
seeds per pod (r = -0.18) showed a negative association with single plant
yield. The path analysis also revealed that the number of pods per plant (0.94)
contributed the highest positive direct effect to single plant yield followed
by the hundred seed weight (0.32). The
Table 3: Qualitative trait-based clustering of TNAU soybean core
set
|
Clusters |
No. of Genotypes |
Genotypes |
|
I |
19 |
KDS343, JS20-01, JS99-72, JS(SH)99-02, JS(SH)99-14,
NRC2007-A-23, NRC77, MACS1184, AMSS44, AMSS463, LU75, CLARK, CSB0804,
CSB0806, Co2, RKS18, MAUS61, JS335 & EC250607 |
|
II |
9 |
NRC2006-m-6, DS2402, TNAU20049, EC325099, UGM75,
AGS747, EC30198, EC73-16E & JS9305 |
|
III |
17 |
RSC14, IC16009, JS98-21, PK1125, IC13051, EC36961,
EC39498, EC62376, AVRDC508, AVRDC576, EC50082, EC799, EC39536, EC4290,
IC109544, Co1 & Co (Soy)3 |
|
IV |
1 |
MAUS59 |
|
V |
2 |
JS20-09 & EC7587 |
|
VI |
1 |
TNAU20048 |
|
VII |
1 |
EC109556 |
%20184-192_files/image008.png)
Fig. 4: Rotated component matrix for the ten principal
components
|
|
|
|
Fig. 5: Correlation between ten quantitative
characters in the TNAU soybean core set |
Fig. 6: Path analysis for ten quantitative
characters in the TNAU soybean core set |
%20184-192_files/image010.png)
%20184-192_files/image012.png)
Table 4: Similarity indices of selected parents for
hybridization based on qualitative traits
|
Diverse parents |
Closely related parents |
||||||||
|
Parents |
JS(SH)99-02 |
IC16009 |
TNAU20048 |
JS9305 |
Parents |
NRC77 |
UGM75 |
EC30198 |
Co(Soy)3 |
|
JS(SH)99-02 |
1 |
|
|
|
NRC77 |
1 |
|
|
|
|
IC16009 |
0.34 |
1 |
|
|
UGM75 |
0.80 |
1 |
|
|
|
TNAU20048 |
0.13 |
0.25 |
1 |
|
EC30198 |
0.75 |
0.87 |
1 |
|
|
JS9305 |
0.31 |
0.40 |
0.29 |
1 |
Co(Soy)3 |
0.7 |
0.76 |
0.89 |
1 |
Table 5: Performance of F2 segregants involving
distantly related soybean accessions
|
Characters |
F2 population |
Parents |
|||||||||||||||||
|
F2(IC16009 × JS(SH)99-02) |
F2(IC16009 × TNAU 20048) |
F2(IC16009 × JS 9305) |
JS(SH) 99-02 |
IC 16009 |
TNAU 20048 |
JS 9305 |
|||||||||||||
|
Mean |
Mini. |
Maxi. |
CD |
CV |
Mean |
Mini. |
Maxi. |
CD |
CV |
Mean |
Mini. |
Maxi. |
CD |
CV |
Mean |
Mean |
Mean |
Mean |
|
|
PH |
65.8 |
32.6 |
85.9 |
8.9 |
7.4 |
75.5 |
54.6 |
94.3 |
11.5 |
9.1 |
80.2 |
61.3 |
91.6 |
12.3 |
7.9 |
15.9 |
87.9 |
46.0 |
26.2 |
|
DFF |
39.2 |
28.0 |
47.0 |
4.2 |
7.9 |
43.7 |
39.0 |
50.0 |
8.7 |
6.9 |
38.5 |
35.0 |
42.0 |
3.4 |
6.3 |
31.0 |
41.7 |
45.7 |
35.0 |
|
DM |
87.4 |
50.0 |
95.0 |
10.5 |
8.0 |
94.0 |
86.0 |
105.0 |
8.9 |
7.8 |
88.6 |
82.0 |
94.0 |
5.1 |
4.2 |
77.7 |
91.7 |
95.0 |
85.0 |
|
NPB |
6.8 |
4.0 |
8.0 |
1.1 |
7.5 |
7.8 |
6.0 |
10.0 |
1.3 |
6.9 |
6.4 |
6.0 |
8.0 |
0.6 |
5.3 |
4.0 |
9.3 |
6.3 |
3.3 |
|
NCP |
90.9 |
59.0 |
128.0 |
15.6 |
15.1 |
91.3 |
66.0 |
91.0 |
9.6 |
12.4 |
52.3 |
46.0 |
89.0 |
8.9 |
8.2 |
25.7 |
85.3 |
16.0 |
13.0 |
|
NPP |
267.2 |
123.0 |
298.0 |
23.1 |
18.2 |
288 |
156.0 |
291.0 |
16.3 |
16.2 |
198.6 |
160.0 |
267.0 |
16.5 |
12.6 |
88.3 |
245.7 |
84.3 |
40.3 |
|
NSP |
2.6 |
2.0 |
3.0 |
0.3 |
2.1 |
2.6 |
2.0 |
3.0 |
0.2 |
3.0 |
3.0 |
2.6 |
3.0 |
0.1 |
3.0 |
3.0 |
2.3 |
2.3 |
3.0 |
|
HSW |
9.2 |
7.6 |
9.9 |
0.8 |
2.8 |
8.9 |
6.9 |
10.2 |
0.7 |
3.1 |
11.2 |
9.9 |
12.4 |
0.7 |
4.1 |
12.1 |
7.1 |
9.7 |
12.3 |
|
SPY |
45.78 |
38.7 |
50.35 |
6.4 |
4.6 |
44.6 |
29.4 |
48.6 |
8.4 |
7.2 |
43.8 |
22.1 |
46.9 |
9.1 |
7.9 |
22.9 |
39.2 |
17.4 |
15.6 |
*CD @
5%
Table 6: Performance of F2 segregants involving
closely related soybean core accessions
|
Characters |
F2 population |
Parents |
|||||||||||||||||
|
F2(Co (Soy)3 × NRC 77) |
F2(Co (Soy)3 × UGM 75) |
F2(Co (Soy)3 × EC 30198) |
NRC 77 |
UGM 75 |
EC 30198 |
Co (Soy)3 |
|||||||||||||
|
Mean |
Mini. |
Maxi. |
CD |
CV |
Mean |
Mini. |
Maxi. |
CD |
CV |
Mean |
Mini. |
Maxi. |
CD |
CV |
Mean |
Mean |
Mean |
Mean |
|
|
PH |
40.6 |
27.3 |
66.8 |
8.6 |
9.6 |
40.6 |
32.7 |
59.6 |
6.7 |
10.8 |
56.2 |
50.6 |
62.1 |
2.8 |
7.6 |
22.3 |
22.4 |
53.6 |
55.6 |
|
DFF |
37.2 |
32.0 |
44.0 |
3.2 |
7.2 |
42.6 |
40.0 |
44.0 |
1.8 |
5.0 |
40.5 |
38.0 |
44.0 |
1.9 |
5.8 |
35.3 |
41.3 |
39.0 |
40.0 |
|
DM |
86.4 |
82.0 |
92.0 |
3.5 |
7.6 |
88.6 |
84.0 |
92.0 |
3.1 |
7.6 |
88.0 |
83.0 |
92.0 |
2.0 |
8.6 |
84.3 |
86.0 |
84.3 |
90.0 |
|
NPB |
6.4 |
5.0 |
7.0 |
0.3 |
5.6 |
6.1 |
5.0 |
7.0 |
0.4 |
5.1 |
6.0 |
5.0 |
7.0 |
0.6 |
7.6 |
6.0 |
5.3 |
5.0 |
6.3 |
|
NCP |
36.8 |
32.0 |
42.3 |
2.6 |
10.2 |
41.3 |
36.0 |
48.0 |
3.9 |
9.4 |
43.0 |
30.0 |
45.0 |
3.1 |
9.8 |
35.3 |
26.0 |
35.3 |
40.0 |
|
NPP |
120.5 |
96.0 |
140.2 |
8.6 |
12.3 |
118.9 |
102.0 |
123.0 |
4.9 |
8.6 |
120.4 |
70.0 |
130.0 |
9.9 |
10.2 |
112.3 |
75.0 |
72.0 |
111.7 |
|
NSP |
2.7 |
2.3 |
3.0 |
0.2 |
4.3 |
2.7 |
2.7 |
3.0 |
0.2 |
3.5 |
2.7 |
2.7 |
3.0 |
0.2 |
4.9 |
2.7 |
2.7 |
2.7 |
2.7 |
|
HSW |
9.9 |
8.5 |
10.6 |
0.9 |
5.6 |
10.2 |
10.0 |
10.6 |
0.1 |
2.6 |
10.2 |
9.9 |
10.6 |
0.3 |
8.6 |
8.4 |
10.2 |
9.6 |
10.4 |
|
SPY |
25.1 |
20.1 |
28.3 |
2.5 |
10.2 |
24.2 |
18.9 |
26.4 |
1.6 |
6.6 |
24.6 |
18.5 |
27.5 |
2.7 |
5.9 |
22.3 |
18.3 |
15.8 |
22.9 |
*CD @
5%
PH–Plant
Height (cm): DM- Days to Maturity: DFF- Days to Fifty percent Flowering: NPB-
Number of Primary Branches: NCP- Number of Clusters per Plant: NPP- Number of
Pods per Plant: NSP- Number of Seeds per Pod: HSW- Hundred Seed Weight: SPY-
Single Plant Yield (g)
negative direct effect on single
plant yield was exerted by days to maturity (-0.09), and number of clusters per
plant (-0.04). The highest positive
indirect effect to single plant yield was observed with number of pods per
plant and number of clusters per plant (0.91) followed by number of clusters
per plant and plant height (0.80). The residual effect was only eight percent.
Qualitative Trait based Experiments
The
variations for qualitative traits categorized the genotypes into eight clusters
(Table 3), the cluster I was the biggest (19 genotypes) followed by cluster III
(17) and cluster II (9). The genotypes MAUS59 and TNAU20048 formed a solitary cluster individually. Further, it
also revealed a significant variation for growth habit, leaf type, hilum colour and pod colour (Fig. 7). A
relationship between hypocotyl and flower colour was
established. The genotypes with
purple hypocotyl produced purple flowers, while the green hypocotyl genotypes
produced white flowers. A total of 38 genotypes were earmarked as determinate
type, 10 and two genotypes were categorized as semi-determinate and
indeterminate respectively. The biochemical based KOH and NaOH tests also
grouped the genotype EC109556 as a solitary cluster. The seed coat peroxidase
test equally distributes the genotypes into two groups based on the presence or
absence of enzyme activity.
The results of quantitative (PCA biplot) (Fig. 2) and
qualitative trait (similarity index) (Table 4) based experiments along with
single plant yield (Table 5 and 6) were
utilized to tag potential genotypes for targeted hybridizations. Accordingly,
two groups of parents were formed. The genotypes located far apart (encircled
as red) with less similarity index and average yield potential (single plant yield above the national check JS335
i.e., > 15.2 g) were categorized as group I (IC16009,
JS(SH)99-02, TNAU20048 and JS9305). While group II comprised of closely related
parents (encircled as green) with higher similarity index and good yield
(Co(soy)3, NRC77, UGM75 and EC30198). Based on single plant the genotype IC
%20184-192_files/image014.png)
Fig. 7: Variations observed for qualitative traits in the TNAU
soybean core set
16009 (39.2 g) and Co (soy) 3 (22.9 g) were designated
as female parents in group I and II respectively while the other three parents
were treated as pollen parents. In each group three crosses were made and the
progenies were evaluated separately. The true F1s were forwarded to
F2 and the agronomic performances were evaluated along with
respective parents (Table 5 and 6).
Validation of
Breeding Value of Soybean Core Set
The above listed distantly and closely related genotypes
utilized for hybridization to validate the results of quantitative and
qualitative trait-based experiments. The genotypes were hybridized, the F1s
were grown in an ideal condition, the true F1s were forwarded to F2
generation, and the yield performances were analyzed. The F2
segregants of closely related accessions showed the average single plant yield
of 25.1
g [F2 (Co(soy)3 × NRC77)], 24.2 g [F2 (Co(soy)3 × UGM75)]
and 24.6 g [F2 (Co(soy)3 × EC30198)] compared to the better parent
Co(soy)3 (22.9 g). Similarly, the diverse accessions possessed the average
single plant yield of 45.78 g [F2 (IC16009 × JS(SH)99-02)], 44.6 g
[F2 (IC16009 × TNAU20048)] and 43.2 g [F2 (IC16009 ×
JS9305)] compared to the better parent IC16009 with 39.2 g (Table 5 and 6).
Discussion
Soybean is one
of the major legume crops that grown in different seasons and environments in
India for multi-purposes. The potential yield is not realized due to the
incidence of various stresses and geographical origin (Malik et al. 2011). Therefore, identification
of adaptable, region and trait-specific genotypes and their subsequent
utilization in the pre-breeding activities are necessitated. A core set reduces
the complexity in the modus operandi
of germplasm. The PCA helps to reduce the complexity of multidimensional data
into fewer principal components (PC’s) without losing major information
(Jolliffe and Cadima 2016). In the present study, the
first three PC’s contributed about 73.72% of total variation and were
considered as significant as their Eigenvalues were more than one. The
Eigenvalues of more than one contributes the maximum to the total variation
while less than one contributes less (Gerrano et al. 2019). The genotypes scattered
along the biplot based on the first two PC’s represented the variability among
the soybean core set. The genotypes JS9305, JS(SH)99-02, IC16009 and TNAU20048
were located far apart and considered as highly divergent lines. The characters
viz., single plant yield, number of
pods per plant, and plant height were away from an origin that contributed the
maximum for the divergence. The contributions of various quantitative
characters to various PC’s revealed that the single plant yield and number of
pods per plant, days to maturity and number of seeds per pod contributed the
maximum for the rotated PC1, PC2 and PC3, respectively. Earlier, a similar
finding was reported by (Lazaridi et al. 2017). The ultimate aim of any
breeding program is to improve the yield which is a complex trait and
influenced by contributing traits. An association study on these traits helps
in ideotype breeding (Jain et al.
2018). The results of correlation and path analyses revealed that number of
pods per plant and number of clusters per plant contributed significantly to
the single plant yield that was parallel with the findings of Jain et al. (2018).
Qualitative traits categorized the genotypes into eight
clusters. The genotypes within a cluster are considered as less divergent while
the genotypes grouped in different clusters are more divergent. The qualitative
trait-based clustering also confirms the grouping pattern of genotypes as per
quantitative traits (Ramteke
and Murlidharan 2012). The determinate growth type
accessions found in the study might be preferred under rainfed conditions as it
conserves moisture (Malik et al.
2011). The accessions that possessed a lanceolate leaflet shape might be linked
to drought tolerance. Similar findings were reported by Malik et al. (2011). Variations in hilum colour and pod colour are due to
light intensity, temperature, drought, disease injury, and other environmental
factors (Yadav and Sharma 2001). The KOH, NaOH and seed coat peroxidase tests
helped in effective grouping in the present study. Earlier, the utility of
biochemical tests in clustering was reported (Agrawal and Sharma 1989).
To validate the above results, four diverse and closely
related parents were selected based on quantitative and qualitative clustering
patterns and hybridized. The F2 generation of closely related and
diverse accessions showed an average yield increase of 7.5 and 14.1 percent,
respectively than their better parent. These results indicated that the TNAU
soybean core set is more diverse and can be effectively used for pre-breeding
activities.
Conclusion
The multivariate analysis for various quantitative and
qualitative traits revealed the existence of significant variation in the TNAU
soybean core set. The crosses between identified diverse accessions have a net
worthy yield advantage over the closely related genotypes. It is concluded that
the variation present in the TNAU soybean core set has breeding potency and
hence could be used in the pre-breeding programs.
Acknowledgements
The authors are
grateful to Dr. D. Packiaraj, Dr. J.R. Kanna Bapu, Dr. K. Ganesamurthy, Dr. V. Thiruvengadam,
Dr. S. Ganesh Ram, Centre for Plant Breeding and Genetics, Tamil Nadu
Agricultural University for their scientific support.
Author
Contributions
RS and CV planned the experiments. EV and VA performed
the experiments, statistically analyzed the data and made illustrations. EV and
RS wrote the draft manuscript. JR, CV and NS interpreted the results and made
the final Manuscript.
Conflicts of
Interest
All authors declare no conflicts of interest.
Data Availability
The datasets generated during the study are all included
in the manuscript. Further inquiries can be directed to the corresponding
author
Ethics Approvals
Ethics approval was not required for this study
References
Agrawal RL, BL Sharma (1989). Identification of mungbean varieties on the basis of seedling growth response
to GA3 and DDT treatment. Seed Res 17:84‒37
Bisen
A, D Khare, P Nair, N Tripathi (2015). SSR analysis
of 38 genotypes of soybean (Glycine max
(L.) Merr.) genetic diversity in India. Physiol Mol Biol Plants 21:109‒115
Dewey
DR, K Lu (1959). A correlation and path-coefficient analysis of components of
crested wheatgrass seed production 1. Agron J 51:515‒518
El-Harty
EH, SS Alghamdi, MA Khan, HM Migdadi, M Farooq
(2018). Adaptability and stability analysis of different soybean genotypes
using biplot model. Intl J Agric Biol
20:2196‒2202
Epskamp S, MS Epskam, S Mplus Automation
(2019). Package ‘semPlot’. R Package Version
Gerrano AS, WSR
Jansen, FR Kutu (2019). Agronomic evaluation and identification of potential
cowpea (Vigna unguiculata L. Walp) genotypes in South Africa. Acta Agric Sci B Soil
Plant Sci 69:295‒303
Jackson JE, A Edward (1991). User’s guide to principal components. John Willey Sons. Inc., New
York, USA
Jain
RK, A Joshi, HR Chaudhary, A Dashora, CL Khatik (2018). Study on genetic variability, heritability
and genetic advance in soybean [Glycine
max (L.) Merrill]. Leg Res
41:532‒536
Jolliffe IT, J Cadima (2016).
Principal Component Analysis: A review and recent developments. Phil Trans Roy Soc A 374:1–16
Lazaridi E, G Ntatsi, D Savvas, PJ Bebeli (2017). Diversity in
cowpea (Vigna unguiculata (L.) Walp.) local populations from Greece. Genet Resour Crop Evol 64:1529‒1551
Lę S, J Josse, F Husson (2008). FactoMineR: An R
package for multivariate analysis. J Stat
Softw 25:1‒18
Malik FA, M Ashraf, AS Qureshi, MR Khan (2011).
Investigation and comparison of some morphological traits of the soybean
populations using cluster analysis. Pak J
Bot 43:1249‒1255
Mohan S, A Sheeba, T Kalaimagal (2019). Genetic diversity and association
studies in Greengram [Vigna radiata (L.) Wilczek]. Leg Res 44:779‒784
Muthamizhan
G, R Sudhagar, V Thiruvengadam,
S Ganeshram, JR KannanBapu
(2016). Formulation of core set in base collections of soybean (Glycine max (L.) Merrill) germplasm
accessions. Adv Life Sci 5:7909‒7917
Pearson
K (1901). I. Mathematical contributions to the theory of evolution. VII. On the
correlation of characters not quantitatively measurable.
Phil Trans Roy Soc A 195:262‒273
Ramteke R, P Murlidharan (2012).
Characterization of soybean (Glycine max)
varieties as per DUS guidelines. Ind J
Agric Sci 82:572‒577
Rao DE, KV Chaitanya (2020). Changes in the antioxidant
intensities of seven different soybean (Glycine
max (L.) Merr.) cultivars during drought. J Food Biochem
44:e13118
Sharma
S, HD Upadhyaya, RK Varshney, CLL Gowda (2013). Pre-breeding for
diversification of primary gene pool and genetic enhancement of grain legumes. Front
Plant Sci 4:309-322
TNAU – Tamil Nadu Agricultural University (2016). Crop Production Guide (CPG), pp:173‒178.
Tamil Nadu Agricultural University and Department of Agriculture, Government of
Tamil Nadu, India
Upadhyaya HD, CLL Gowda, KN Reddy, S Singh (2009).
Augmenting the pearl millet core collection for enhancing germplasm utilization
in crop improvement. Crop Sci 49:573‒580
Vishwanath,
R Hunje, P Shinde (2021). Influence of Seed Quality
Upgrading Processing Machines in Soybean var. DSb-21. Leg Res 1:1–5
Wei
T, V Simko, M Levy, Y Xie,
Y Jin, J Zemla (2017).
Package ‘corrplot’. Statistician 56:1–18
Yadav S (2016). Soybean (Glycine max L.) germplasm evaluation through morphological and
quality characterization. Intl J Appl
Pure Sci Agric 12:141‒145
Yadav SP, SP Sharma (2001). Variation for hilum colour and its stability during four crop seasons in
soybean (Glycine max). Ind J Agric Sci 110:45‒51