Journal of Genetics

, Volume 92, Issue 3, pp 545–557 | Cite as

Assessment of genetic diversity in Indian rice germplasm (Oryza sativa L.): use of random versus trait-linked microsatellite markers

  • SHEEL YADAV
  • ASHUTOSH SINGH
  • M. R. SINGH
  • NITIKA GOEL
  • K. K. VINOD
  • T. MOHAPATRA
  • A. K. SINGH
Research Article

Abstract

Assessment of genetic diversity in a crop germplasm is a vital part of plant breeding. DNA markers such as microsatellite or simple sequence repeat markers have been widely used to estimate the genetic diversity in rice. The present study was carried out to decipher the pattern of genetic diversity in terms of both phenotypic and genotypic variability, and to assess the efficiency of random vis-à-vis QTL linked/gene based simple sequence repeat markers in diversity estimation. A set of 88 rice accessions that included landraces, farmer’s varieties and popular Basmati lines were evaluated for agronomic traits and molecular diversity. The random set of SSR markers included 50 diversity panel markers developed under IRRI’s Generation Challenge Programme (GCP) and the trait-linked/gene based markers comprised of 50 SSR markers reportedly linked to yield and related components. For agronomic traits, significant variability was observed, ranging between the maximum for grains/panicle and the minimum for panicle length. The molecular diversity based grouping indicated that varieties from a common centre were genetically similar, with few exceptions. The trait-linked markers gave an average genetic dissimilarity of 0.45 as against that of 0.37 by random markers, along with an average polymorphic information constant value of 0.48 and 0.41 respectively. The correlation between the kinship matrix generated by trait-linked markers and the phenotype based distance matrix (0.29) was higher than that of random markers (0.19). This establishes the robustness of trait-linked markers over random markers in estimating genetic diversity of rice germplasm.

Keywords

germplasm genetic diversity microsatellite markers principal component analysis 

Introduction

Rice (Oryza sativa L.) is one of the world’s most widely cultivated crop species. As a major cereal crop, it is one of the most diversified crop species due to its adaptation to a wide range of geographical, ecological and climatic regions. India is remarkably rich in rice diversity, including cultivars, landraces, wild and weedy relatives (DRR, Hyderabad). The rice germplasm is a rich reservoir of valuable genes that plant breeders can harness for crop improvement.

There are several ways for estimation of diversity in germplasm, such as evaluation of phenotypic variation, biochemical and DNA polymorphisms. However, both phenotypic and biochemical characterizations are unreliable because they are environmentally challenged, labour demanding, numerically and phenologically limited. On the contrary, DNA-based molecular markers are ubiquitous, repeatable, stable and highly reliable (Ford-Lloyd et al.1997; Virk et al.2000; Song et al.2003). Among the several classes of available DNA markers, microsatellite or simple sequence repeat (SSR) markers are considered the most suitable due to their multiallelic nature, high reproducibility, codominant inheritance, abundance and extensive genome coverage. A large number of SSR markers have been developed and mapped in rice (Temnykh et al.2000; McCouch et al.2002), which vary in the degree of polymorphism depending on their position in the coding or noncoding segments, nature of their repeat motifs and the genomewide abundance. Therefore, an ideal set of SSR markers providing genomewide coverage will facilitate an unbiased assay of genetic diversity which in turn will provide a robust, unambiguous molecular description of rice cultivars.

The assessment of diversity of Indian rice germplasm has been carried out by many researchers in the past (Saini et al.2004; Ram et al.2007; Sundaram et al.2008; Kumar et al.2010; Sivaranjani et al.2010; Vanniarajan et al.2012). However, most of these studies have used a set of random, nondescript SSR markers. It has been hypothesized that the use of random markers for assessing genetic diversity might not reflect the functionally useful variations prevalent at the coding regions of the genome (Zhang et al.2010), a crucial requisite for the breeding programmes, where the diversity occurring at identified ‘heterotic loci’ should be taken into account for selection of suitably diverse parental lines. Therefore, it is pertinent to study and compare the pattern of genetic diversity, as established, by using random vis-à-vis trait-linked SSR markers, which would confirm their suitability to assess genetic diversity.

In this investigation, we selected 88 genotypes that included landraces, mega-varieties, elite and popular cultivars, and breeding lines developed and sourced from different rice breeding centres across India having common knowledge on their variability, adaptability and popularity with the objective of identifying genotypic heterotic pools to exploit heterosis and recombination advantage in future breeding programmes. Special selection emphasis was given to yield and yield components. While determining genetic diversity of 88 rice germplasm accessions, we compared the efficacy of a random set of SSR markers as against another set of SSR markers reportedly linked to yield and yield components.

Materials and methods

Plant materials

Study materials consisted of 88 rice accessions representing landraces, released varieties, cultivars and breeding lines from different breeding centres across India (table 1). The accessions were field-grown during kharif season of 2010 at Indian Agricultural Research Institute, New Delhi, India, in randomized blocks with three replications. The seedlings were raised through panicle to row method to maintain genetic purity. Three rows of four metre each formed the plot for each genotype. The experiment was subjected to standard agronomic management (includes aspects regarding plant geometry, fertilizer application and insect management. These have been standardized for each crop and were implemented here for rice).
Table 1

Details of 88 genotypes used in the study.

 

Variety (other name)

Parentage

Breeding/source location

1

Abhaya

CR157-392/OR57-21

IGKV, Raipur

2

ARS36

-

UAS, Bangalore

3

Bamleshwari (IET14444)

RP2151-40-1/IR9828-23-1

IGKV, Raipur

4

Basmati 370

Pureline selection from local

RRI, Kala Shah Kaku

Dehraduni basmati landraces

5

Basmati 386

Selection from Pakistan Basmati

RRI, Kala Shah Kaku

6

Basmati 564 (RR564)

Sel from Basmati 370

SKUAST, Jammu

7

BhriguDhan (HPR179)

Chucheng/ Daval // Matali

CSKHPKV, Palampur

8

Chandrahasini (IET16800; R979-528-2-1)

Abhaya / Phalguna

IGKV, Raipur

9

China 988

Introduction from China

CSKHPKV, Palampur

10

Chinikamini

Landrace

OUAT, Bhubaneswar

11

Danteshwari (IET15450)

Samridhi/ IR8608-298

IGKV, Raipur

12

Dhanaprasad

Landrace

OUAT, Bhubaneswar

13

Govind

IR20/ IR24

GBPUAT, Pantnagar

14

Hassan Serai

Introduction from Iranian Basmati

CSKHPKV, Palampur

15

Himalaya 1

IR8/ Tadukan

CSKHPKV, Palampur

16

Himalaya 741

CR125-42-5/ IR2061-213

CSKHPKV, Palampur

17

Himalaya 799 (HPU799)

IR28/ Shensi Var// IR28

CSKHPKV, Palampur

18

Himdhan (RxT42)

R575/TN1

CSKHPKV, Palampur

19

HPR1068

IR42015-83-3-2-2/ IR9758-K2

CSKHPKV, Palampur

20

HPR2143

PhulPatas 72/ HPU741

CSKHPKV, Palampur

21

HUBR2-1 (MalviyaBasmati 1)

HBR92/ Pusa Basmati/ Kasturi

BHU, Varanasi.

22

HUR105 (MalviyaSugandh 105)

Mutant of MPR 7-2

BHU, Varanasi.

23

HUR3022 (MalviyaDhan 2)

IR 36/ HR 137

BHU, Varanasi.

24

HUR36 (MalviyaDhan 36)

Mutant of Mahsuri

BHU, Varanasi.

25

HUR4-3 (MalviyaSugandh 4-3)

Mutant of Lanjhi

BHU, Varanasi.

26

Indira Sona

IR58025A/ R710-437-1-1

IGKV, Raipur

27

Indira SugandhitDhan 1

Madhuri/ Surekha

IGKV, Raipur

28

IR24

IR8/ IR127-2-2

IRRI, Los Baños

29

IR64

IR5657-33-2-1/ IR2061-465-1-5-5

IRRI, Los Baños

30

Jaiphulla

Landrace

OUAT, Bhubaneswar

31

Jaldubi (IET 17153; AR1023)

Sel. Surguja local

IGKV, Raipur

32

JR201 (Rashmi)

IR36/ JR75

JNKV, Jabalpur

33

JR75

IR20/ L14// BSJ205

JNKV, Jabalpur

34

K332

Shinei/ Marin11

SKUAST, Srinagar

35

K429 (Kohsaar)

Shinei/ Ginmsari

SKUAST, Srinagar

36

Karma Mahsuri

Mahsuar/ R 296-260

IGKV, Raipur

37

Kasturi

Basmati 370/ CR88-17-1-5

CSKHPKV, Palampur

38

Kranti

Cross 116/ IR8

JNKV, Jabalpur

39

Mahamaya

Asha/ Kranti

IGKV, Raipur

40

MalaviyaDhan

Mutant of Mahsuri

BHU, Varanasi

41

Manhar

IR24/ Cauvery

GBPUAT, Pantnagar

42

MAS109

IR64/ Azucena

UAS, Bangalore

43

MAS25

IR64/ Azucena

UAS, Bangalore

44

MAS868

IR64/ Azucena

UAS, Bangalore

45

MAS946

IR64/ Azucena

UAS, Bangalore

46

MAS946-1 (Sharada)

IR64/ Azucena

UAS, Bangalore

47

MR219

MR137/ MR151

MARDI, Selangor

48

MR220

MR137/ MR151

MARDI, Selangor

49

NaggarDhan

ChingSai 25

CSKHPKV, Palampur

50

OYR128

UAS, Bangalore

51

OYR69

UAS, Bangalore

52

PalamDhan 957

IR32429-122-3-1-2/IR31851-64-2-3-2

CSKHPKV, Palampur

53

Pant Dhan 10

IR32/ Mahsuri// IR28

GBPUAT, Pantnagar

54

Pant Dhan 11

VL206/ Dagi

GBPUAT, Pantnagar

55

Pant Dhan 12 (IET10955)

Govind/ UPR201-1-1

GBPUAT, Pantnagar

56

Pant Dhan 15

Basmati 370/ Sudari/ Behral/ Muskan 41

GBPUAT, Pantnagar

57

Pant Dhan 16

BG380/ BG367-4

GBPUAT, Pantnagar

58

Pant Dhan 18

IR25393-57/ RD23// IR27316-96///SPRLR77205-3-2/ SPRLR79234-51-2

GBPUAT, Pantnagar

59

Pant Dhan 19

BG132/ UPRI95-141

GBPUAT, Pantnagar

60

Pant Dhan 6

IR8608-298-3-1/ IR10179-23

GBPUAT, Pantnagar

61

Pant SankarDhan 1

UPR195-178A/ UPR192-133R

GBPUAT, Pantnagar

62

Pant SankarDhan 3

UPRI95-17A/ UARI93-287 R

GBPUAT, Pantnagar

63

Pant SugandhDhan 17

Pusa Basmati 1/UPRM500

GBPUAT, Pantnagar

64

PAU201

PR103 / PAU1126

PAU, Ludhiana

65

Poornima

Poorva/ IR8608-298

JNKV, Jabalpur

66

PR111 (IET13576)

IR54/ PR106

PAU, Ludhiana

67

PR114

TN1/ Patong 32// PR106*4/// IR8

PAU, Ludhiana

68

PR115

RP2151-173-1-8/ PR103*3

PAU, Ludhiana

69

PR116

PR108/// TN1/ Patong 32// PR106*6//// PR108

PAU, Ludhiana

70

PR118

Pusa 44/ PR110// Pusa 44*3

PAU, Ludhiana

71

PR120

PAU1196-14-2-5-1-3/ SR817-255

PAU, Ludhiana

72

Prasad

IR747-B-26-3/ IR57948

GBPUAT, Pantnagar

73

Punjab Basmati 2

Basmati 386/ Super Basmati

PAU, Ludhiana

74

Ranbir Basmati (IET11348)

Sel from Basmati 370-90-95

SKUAST, Jammu

75

Saanwal Basmati (IET15815)

Sel from Basmati 370

SKUAST, Jammu

76

SambhaMahsuri (BPT5204)

GEB24/ TN1// Mashuri

ANGRAU, Bapatla

77

Samleshwari (IET17455; R1027-2282-2-1)

R310-37/ R308-6

IGKV, Raipur

78

Sarjoo 52

TN1/ Kashi

NDUAT , Faizabad

79

Shyamala

R60-2713/ R238-6

IGKV, Raipur

80

SKAU23 (Chenab)

K21-9-10-1/ IR2053-521-1-2

SKUAST, Srinagar

81

SKAU27 (Jhelum)

Jinoku/ IET4444

SKUAST, Srinagar

82

SKAU5 (K39)

China 1039/IR580-19-2-3-3

SKUAST, Srinagar

83

SonaMahsuri (BPT 3291)

Sona/ Mashuri

ANGRAU, Bapatla

84

SR1 (Shalimar Rice 1)

CH1007/ IET1444

SKUAST, Srinagar

85

SukaraDhan 1 (HPR1156)

IR32429-122-3-1-2/ IR31868-64-2-3-3-3

CSKHPKV, Palampur

86

Super Basmati

IR662/Bamati 320

RRI, Kala Shah Kaku

87

Swarna (MTU7029)

Vasista / Mahsuri

APRRI, Maruteru

88

T23

Sel from Kala Sukhdas

CSKHPKV, Palampur

JNKV, Jawaharlal Nehru Krishi Vishvavidyalaya; OUAT, Orissa University of Agriculture and Technology; BHU, Banaras Hindu University; UAS, University of Agricultural Sciences; PAU, Punjab Agricultural University; SKUAST, Sher-e-Kashmir University of Agricultural Sciences and Technology; RRI, Rice Research Institute; IGKV, Indira Gandhi Krishi Vishwavidyalaya; CSKHPKV, Chaudhary Sarwan Kumar Himachal Pradesh Krishi Vishvavidyalaya; GBPUAT, Govind Ballabh Pant University of Agriculture and Technology; IRRI, International Rice Research Institute; ANGRAU, Acharya NG Ranga Agricultural University; APRRI, Andhra Pradesh Rice Research Institute; MARDI, Malaysian Agricultural Research and Development Institute; NDUAT, Narendra Dev University of Agriculture and Technology; IARI, Indian Agricultural Research Institute

Evaluation of morphological traits

Data were recorded on five competitive plants per genotype from the middle row for yield and yield components such as days to 50% flowering (DFL), days to maturity (DMT), plant height (PLH), number of tillers per plant (TLN), panicle length (PNL), single plant yield (SPY), 1000-grain weight (TGW), total grains per panicle (TGN), number of filled grains per panicle (FGN), number of unfilled grains per panicle (UGN) and spikelet fertility (SFP).

Selection of SSR markers

A priori information was used to select random as well as trait-linked SSR markers. Selected random markers consisted of a set of 50 markers having genomewide distribution that could effectively illustrate the diversity among the germplasm, shortlisted by the International Rice Research Institute under the Generation Challenge Program (http://gramene.org/markers/microsat/50_ssr.html). Another set of 50 SSR markers linked to yield and related traits were selected based on the mapping studies conducted and the results that have been documented hitherto in rice (Zhang et al.2010). However, there was a perceptible limitation with the latter as they did not provide genomewide coverage.

SSR marker assay

Total genomic DNA isolation was carried out using a microextraction method (Prabhu et al.1998) and assayed with a total of 100 SSR markers. The PCR mixture contained 25–30 ng template DNA, 5 pmol of each primer, 0.05 mM dNTPs, 10× PCR buffer and 0.5 U of Taq DNA polymerase in a reaction volume of 10 μL. Template DNA was initially denatured at 94°C for 5 min followed by 35 cycles (1 min denaturation at 94°C, 1 min annealing at 55°C and 2 min of primer extension at 72°C) of PCR amplification followed by final extension of 72°C for 7 min. The amplification products were separated on 3.5% metaphor agarose gels and run for 3 h in 1× TAE buffer. DNA fragments were visualized under UV transillumination using Bio-Rad Molecular Imager (GelDocTM XR+ Imager, New Delhi, India).

Data analysis

The phenotypic data were subjected to statistical analysis for descriptive purposes, as well as for character association. Since the data contained multicollinear variables, a principal component analysis (PCA) based clustering was performed. The genotypic factor scores corresponding to significant principal components (PCs) were used to perform cluster analysis using pair group distance with Euclidean similarity measure. The dendrogram was constructed using the software package PAST (paleontological statistics software package for education and data analysis, Hammer et al. (2001).

Amplicons were scored for their product size for each primer genotype combination. The informativeness of the markers in terms of polymorphism was determined by calculating the polymorphism information content (PIC, Botstein et al. 1980) for each SSR locus, according to the formula:
$$ \mbox{PIC}=\left( {1-\sum\limits_{i=1}^k {\hat{p}_i^2 } } \right)\frac{2n}{2n-1}-\sum\limits_{i=1}^{k-1} {\sum\limits_{j=i+1}^k {2\hat{p}_i^2 \hat{p}_j^2 } } , $$
where \(\hat{p}_i \) was the estimated allele frequencies of k alleles (i = 1 to k) and n, the number of individuals sampled. The allelic data was analysed using the computer package Darwin (ver. 5.0.158) (Perrier and Collet 2006). The dissimilarity coefficients were calculated between genotypes using simple matching coefficient (dij).
$$ \mbox{d}_{ij} =1-\frac{1}{L}\sum\limits_{l=1}^L {\frac{m_l }{\pi }}. $$
Where L is the total number of loci (l = 1 to L), mi is the number of matching alleles for locus l, and π is the ploidy level.

The dissimilarity matrix was used for clustering of genotypes, based on unweighted neighbour-joining method. Confidence limits of different clades were tested by bootstrapping 10,000 times to assess the repetitiveness of genotype clustering (Felsenstein 1985).

To study relative efficiency of diversity assessment of two sets of molecular markers, relative kinship coefficients between the genotypes were estimated as pair-wise dissimilarity for random, gene-linked and pooled marker data sets. The Euclidean distance matrix from the genotypewise factor loadings for the significant PCs was also used in comparison. A Mantel test was done to test similarities of distance and kinship matrices, with 30,000 permutations for two-tailed test. Mantel test was performed using the software package PASSaGE (Rosenberg and Anderson 2011).

Results

Morphological characterization

Evaluation of morphological traits in the rice germplasm lines revealed that wide range of variation existed among the genotypes, for traits such as UGN, SPY, FGN, TLN and TGN showing high coefficients of variation (table 2) and SPF, PNL and DMT showing the lowest. The earliest flowering line was SKAU 5 (79 days) while flowering in Chinikamini took as long as 159 days. The average TLN varied from 8.8 in Super Basmati to 42.6 for BhriguDhan. The SPY was highest for MAS 25 (43.0 g), while it was lowest in Jaldubi (4.5 g). Pant Dhan 19 had the maximum TGW (31.1 g) whereas it was least for Chinikamini (8.5 g). The SPF of the accessions ranged from 38.2% in HPR 2143 to 95.7% in Mahamaya (table 1 in electronic supplementary material at http://www.ias.ac.in/jgenet/). Pearson’s correlation coefficients between the phenotypic traits are given in table 3. There were some highly correlated characters, such as DMT and DFL (0.99) and FGN and TGN (0.91), which may bias the estimates of diversity pattern due to multicollinearity. Therefore, an independent estimate of variability was performed by a PCA.
Table 2

Variability observed for agronomic traits.

Trait

Mean (±SE)

Range

CV (%)

Days to 50% flowering (DFL)

104.73 (2.11)

79.0–159.0

18.89

Days to maturity (DMT)

134.61 (2.11)

109.0–189.0

14.7

Plant height (cm) (PLH)

116.60 (2.37)

84.1–177.5

19.07

Number of tillers (TLN)

14.06 (0.49)

8.8–42.6

32.4

Panicle length (cm) (PNL)

26.93 (0.36)

18.8–36.3

12.37

Yield/plant (g) (SPY)

25.29 (0.93)

4.5–43.0

34.32

Test weight (g) (TGW)

21.60 (0.43)

8.5–31.1

18.67

Total grains per panicle (TGN)

172.6 (5.61)

72.4–326.8

30.47

Filled grains per panicle (FGN)

139.88 (5.00)

60.8–292.6

33.55

Unfilled grains per panicle (UGN)

32.43 (2.33)

6.4–126.2

67.27

Spikelet fertility (%) (SPF)

81.06 (1.04)

38.2–96.7

12.07

Table 3

Phenotypic correlation coefficients among the traits.

 

DFL

DMT

PLH

TLN

PNL

SPY

TGW

TGN

FGN

UGN

DMT

0.999

         

PLH

0.426

0.421

        

TLN

0.225

0.219

−0.143

       

PNL

0.209

0.208

0.504

−0.052

      

SPY

0.106

0.106

−0.048

0.047

0.088

     

TGW

0.412

0.416

−0.180

−0.075

−0.095

0.279

    

TGN

0.608

0.603

0.212

−0.174

0.230

0.243

0.446

   

FGN

0.535

0.531

0.155

−0.142

0.123

0.429

0.318

0.909

  

UGN

0.317

0.312

0.169

−0.108

0.304

0.323

0.391

0.470

0.066

 

SPF

0.005

0.005

−0.045

0.047

−0.182

0.477

0.213

−0.018

0.384

0.828

Critical values of Pearson’s correlation coefficient at df=86, 0.2096 (P = 0.05) and 0.2732 (P = 0.01).

DFL, days to 50% flowering; DMT, days to maturity; PLH, plant height in cm; TLN, number of tillers per plant; PNL, panicle length in cm; SPY, single plant yield in g; TGW, 1000-grain weight in g; TGN, total grains per panicle; FGN, number of filled grains per panicle; UGN, number of unfilled grains per panicle; SFP, spikelet fertility per cent. Significant values are given in bold.

The active characters with their eigenvalues obtained from the PCA of the correlation matrix are given in table 4. There were four major PCs that accounted for the significant part of total variation (78.3%). The first and the second PCs explained 57.2% of the total variation. The variable contributions to the first two PCs are given in figure 1. It is seen that all traits except for SPF, SPY, FGN and UGN significantly contributed to the PC1. SPF and SPY had significant contribution to PC2, while UGN and FGN contributed significantly to the both the components.
Table 4

Eigenvalues of correlation matrix and related statistics for the agronomic traits.

Eigenvalue

Cumulative eigenvalue

Per cent total variance

Cumulative variance %

3.90

3.90

35.48

35.48

2.38

6.29

21.68

57.16

1.28

7.57

11.64

68.79

1.04

8.61

9.49

78.28

0.97

9.58

8.85

87.13

0.65

10.24

5.94

93.06

0.39

10.63

3.52

96.58

0.32

10.99

2.87

99.45

0.06

10.99

0.53

99.98

0.00

10.99

0.02

99.99

0.00

11.00

0.01

100.00

Figure 1

Contribution of the phenotypic traits towards first two PCs.

Contributions of genotypes to major PCs are given in table 5. Highest contribution of variability to the first PC was observed for Dhanaprasad (11.1%), while HPR2143 (15.8%) and Jaldubi (10.9%) showed significant contribution towards PC2. Two dimensional scaling of the genotypes by the first two PCs (figure 2) showed two distinct groups of genotypes, first with Dhanaprasad, Jaldubi and HPR2143, and the second with rest of all other genotypes. Among the second and the largest group, there were many significant variants, such as Jaiphulla, Indira Sona, MalaviaDhan, Chinikamini, Abhaya, Mahamaya, Swarna, Samba Mashuri and NaggarDhan that had either significant contributions to either PC1 or PC2 or both. Further, the cluster analysis based on the unweighted pair group averaging of Euclidean distances of the factor ladings of all four significant PCs revealed two major clusters at 100% level and four clusters at 80% level (figure 3). The two major clusters correspond to those identified by the two-dimensional scaling between first two PCs. At 80% level, HPR2143 formed a separate cluster, followed by Dhanaprasad and Jaldubi in second cluster. BhirguDhan 1 formed another single member cluster, followed by a large cluster of all the remaining entries.
Table 5

Contributions of genotypes in percentage to major PCs based on correlations.

Genotypes

PC1

PC2

PC3

PC4

Abhaya

0.32

2.58

1.24

0.01

ARS36

0.01

1.07

0.64

0.00

Bamleshwari

0.54

2.67

0.91

1.73

Basmati370

0.92

0.03

6.39

0.27

Basmati386

0.29

0.20

5.59

0.20

Basmati564

0.10

0.01

7.79

0.16

BhirguDhan1

3.40

0.01

1.57

18.98

Chandrahasini

0.13

0.40

0.00

0.28

China988

0.47

0.18

2.29

0.86

Chinikamini

5.24

0.04

1.41

1.06

Danteshwari

0.67

0.23

0.78

0.25

Dhanaprasad

11.10

8.67

0.28

0.46

Govind

1.71

0.91

0.19

0.06

Hassanserai

0.33

1.20

1.72

0.33

Himalaya1

0.59

0.19

2.10

0.35

Himalaya741

1.57

0.10

0.67

0.71

Himalaya799

3.62

0.00

0.32

1.10

Himdhan

0.76

0.04

0.97

0.18

HPR1068

2.46

0.00

0.18

0.67

HPR2143

0.01

15.77

0.07

2.24

HUBR2-1

0.00

0.21

0.14

0.37

HUR105

0.09

0.09

0.77

1.13

HUR302

0.03

0.97

0.10

0.07

HUR36

3.10

2.58

0.72

0.00

HUR4-3

0.01

0.03

0.27

0.28

IndiraSona

1.48

1.55

0.02

3.01

IndiraSugandhitDhan1

0.26

0.30

0.25

1.89

IR24

0.26

0.00

0.24

0.90

IR64

0.22

0.34

0.00

0.36

Jaiphulla

6.46

1.11

3.13

0.36

Jaldubi

6.80

10.92

0.08

1.47

JR201

2.26

0.01

0.15

0.49

JR75

0.96

3.61

0.37

0.61

K332

2.40

0.02

0.09

1.65

K429

2.76

0.09

1.80

1.65

KarmaMahsuri

0.46

0.34

1.65

0.78

Kasturi

0.32

0.01

5.18

0.01

Kranti

0.13

0.15

0.16

0.50

Mahamaya

0.82

2.80

1.00

1.00

MalviyaDhan

4.52

0.17

0.55

0.45

Manhar

0.26

1.89

0.06

0.23

MAS109

0.97

0.17

1.93

0.02

MAS25

0.09

2.13

0.51

8.70

MAS868

0.44

0.83

2.74

0.03

MAS946

0.12

0.11

0.63

0.02

MAS946-1

0.00

0.14

0.19

0.09

MR219

0.22

2.10

0.00

0.04

MR220

0.21

2.24

0.08

0.02

NaggarDhan

3.67

0.23

0.12

1.84

OYR128

0.01

0.83

0.82

1.89

OYR69

0.11

0.29

1.06

3.18

PalamDhan957

1.26

0.02

0.36

0.02

PantDhan10

0.01

2.37

1.39

0.88

PantDhan11

1.22

0.00

0.01

0.48

PantDhan12

0.22

0.23

0.72

0.80

PantDhan15

0.50

0.00

1.49

0.65

PantDhan16

0.23

0.01

0.40

0.20

PantDhan18

0.14

2.11

1.85

0.07

PantDhan19

0.13

0.68

0.18

1.88

PantDhan6

0.95

0.10

1.16

0.24

PantSankarDhan1

0.00

0.18

0.00

0.07

PantSankarDhan3

0.01

0.27

0.45

0.00

PantSugandhDhan17

1.78

0.16

2.04

2.88

PAU201

0.01

1.04

0.05

0.36

Poornima

0.92

0.80

0.24

0.31

PR111

0.03

0.68

0.62

4.03

PR114

0.00

0.10

0.49

0.03

PR115

0.01

0.24

0.00

0.68

PR116

0.35

0.01

1.04

0.84

PR118

0.07

2.33

0.37

1.26

PR120

0.14

0.87

0.09

0.08

Prasad

0.18

1.53

0.10

2.25

Punjab Basmati2

0.11

0.14

1.21

0.02

Ranbir Basmati

0.85

0.67

0.65

3.28

Saanwal Basmati

0.34

0.17

2.18

1.12

Sambha Mahsuri

2.77

3.30

3.41

0.07

Samleshwari

0.02

0.01

0.95

1.71

Sarjoo52

0.12

1.78

1.88

0.70

Shyamala

0.21

0.32

0.04

4.10

SKAU23

1.72

0.09

0.42

0.00

SKAU27

1.25

0.54

0.05

0.44

SKAU5

0.81

2.56

0.12

0.89

SonaMahsuri

0.94

1.97

6.43

0.38

SR1

0.75

0.45

3.28

0.48

SukaraDhan1

1.49

0.14

0.08

0.06

SuperBasmati

0.14

0.00

1.15

2.22

Swarna

4.67

3.53

4.31

0.09

T23

2.02

0.07

0.87

3.53

Figure 2

Two-dimensional scaling of 88 rice genotypes by PCA.

Figure 3

Unweighted pair group average dendrogram of 88 rice genotypes based on Euclidean distances.

Molecular characterization

Random SSR markers

The molecular diversity pattern of random SSR markers showed allelic variation ranging from 1 to 7 (figure 4). Forty-six out of 50 markers were polymorphic. The PIC values of the polymorphic markers ranged from 0.02 (RM 489) to 0.74 (RM 152). The average PIC for these markers was 0.41. The marker RM413 produced a maximum of seven alleles. The average number of polymorphic alleles was 3.63 (table 2 in electronic supplementary material).
Figure 4

A representative gel picture showing allelic diversity for different SSR markers; M, 100-bp ladder; 1–24, rice germplasm lines; A, B, C, D and E, different alleles amplified.

Clustering of the germplasm based on 46 polymorphic markers produced two major clusters at the dissimilarity coefficient of 0.57, and further into four clusters at dissimilarity coefficient of 0.60 (figure 5a). The first cluster consisted of the landraces, japonica varieties and basmati lines. The second and third cluster consisted primarily of indica lines. Pant Dhan 15 and HPR2143 formed the fourth cluster.
Figure 5

Dendrogram of 88 genotypes colour coded against their location of origin (a) as revealed by random markers (b) as revealed by gene linked/based markers.

The average genetic diversity depicted by these markers was 0.37. It was observed that the germplasm/varieties from Bangalore were the least diverse whereas the varieties from Srinagar exhibited maximum genetic diversity (table 6).
Table 6

Genetic dissimilarity values calculated from random SSR markers.

Centre

Average GD

Maximum genetic distance

Minimum genetic distance

Between genotypes

GD

Between genotypes

GD

Overall

0.37

Jaiphulla

Bhirgu Dhan

0.77

MR 219

MR 220

0.01

UAS, Bangalore

0.23

MAS- 868

OYR 128

0.28

MAS-946

MAS-946-1

0.01

PAU, Ludhiana

0.47

Punjab Basmati-2

PAU201

0.70

Punjab Basmati-2

Basmati 386

0.04

SKUAS&T, Srinagar

0.50

SR1

K332

0.60

SKAU 5

SKAU 27

0.07

JNKVV, Jabalpur

0.30

JR201

Poornima

0.35

JR75

Kranti

0.14

OUAT, Bhubaneswar

0.32

Dhanaprasad

Jaiphulla

0.51

Dhanaprasad

Chinikamini

0.19

IGKV, Raipur

0.28

Indira Sugandh Dhan 1

Samleshwari

0.24

Abhaya

Indira Sona

0.18

CSHPKV, Palampur

0.45

Palam Dhan 957

Bhirgu Dhan

0.58

Himalaya 1

Himalaya 741

0.10

GBPUAT, Pantnagar

0.39

Pant Dhan 10

P. Sankar Dhan 1

0.40

Govind

Manhar

0.19

BHU, Varanasi

0.35

HUBR-2-1

HUR - 302

0.32

HUR-36

HUR-302

0.10

GD, genetic distance between two genotypes as determined by the genetic dissimilarity matrix.

SSR Markers for functional genes

Forty-nine SSR markers out of 50 were found to be polymorphic. The average PIC of these markers was 0.48, with the functional marker for Monoculm 1 (Moc1) on chromosome 6 for number of tillers, exhibiting the highest PIC of 0.73. The average number of polymorphic alleles amplified by these markers was 3.67 with RM252 producing a maximum of six alleles (table 3 in electronic supplementary material).

The lines were grouped into two major clusters at the dissimilarity coefficient of 0.55 and further into four clusters at a dissimilarity coefficient of 0.58. The first cluster consisted of the landraces, basmati lines and japonica varieties. The second cluster consisted of varieties from Kashmir and Palampur. The third cluster comprised of varieties from UAS and PAU. The fourth cluster consisted of Pant Dhan varieties (figure 5b).

These markers revealed a higher magnitude of genetic diversity among the germplasm lines. The average genetic diversity was 0.45, with the germplasm from Bangalore being the least diverse and the varieties Srinagar and Palampur being the most diverse (table 7).
Table 7

Genetic dissimilarity values calculated from SSR markers for functional genes.

Centre

Average GD

Maximum genetic distance

Minimum genetic distance

Between genotypes

GD

Between genotypes

GD

Overall

0.45

PR118

K429

0.81

MR 219

MR 220

0.02

UAS, Bangalore

0.20

MAS- 946-1

OYR69

0.31

MAS-946

MAS-946-1

0.04

PAU, Ludhiana

0.50

PR118

Super Basmati

0.76

Basmati 386

P. Basmati 2

0.02

SKUAS&T, Srinagar

0.58

SKAU23

K429

0.75

K332

K429

0.09

JNKVV, Jabalpur

0.40

JR201

Poornima

0.50

JR75

Kranti

0.27

OUAT, Bhubaneswar

0.38

Dhanaprasad

Jaiphulla

0.59

Dhanaprasad

Chinikamini

0.19

IGKV, Raipur

0.40

I. Sugandh Dhan 1

Jaldubi

0.37

Danteshwari

Chandrahasini

0.26

CSHPKV, Palampur

0.54

Palam Dhan 957

Naggar Dhan

0.71

Himalaya 1

Himalaya 741

0.19

GBPUAT, Pantnagar

0.45

Pant Dhan 15

Pant Dhan 19

0.54

Pant Sankar Dhan 3

Pant Sankar Dhan 1

0.12

BHU, Varanasi

0.40

HUR-4-3

HUR 36

0.29

HUR-4-3

HUR-302

0.25

Correlation between genotypic and phenotypic data

The correlation value was calculated from (88 × 88) distance matrices of phenotypic and genotypic datasets using the Mantel test. A moderate but significant correlation (0.29) was observed between kinship matrices of trait-linked markers and phenotype distance matrix. This was higher than the correlation between the kinship matrix of random markers and phenotypic distance matrix (table 8). This indicates that trait-linked markers are more accurate and useful in studying the diversity existing in rice germplasm.
Table 8

Mantel correlations rM (upper diagonal) and two-tailed P values for 30000 permutations (Lower diagonal) between phenotype distance and kinship coefficient matrices for 88 rice genotypes using different sets of marker data.

 

Estimated Mantel correlations (rM)

Kfunc

Krand

Kpooled

Dphen

Kfunc

0.749

0.948

0.290

Krand

0.000

0.921

0.189

Kpooled

0.000

0.000

0.261

Dphen

0.000

0.006

0.003

Kfunc, kinship matrix of trait-linked markers; Krand, kinship matrix of random (diversity panel) markers; Kpooled, kinship matrix of pooled markers, Dphen, phenotype distance matrix.

Discussion

The Indian subcontinent has a very rich diversity in rice germplasm which includes landraces, wild Oryza species, natural hybrids between the cultivars and wild relatives, and the germplasm resources generated in the breeding programmes (Rai 1999). The primary objective of the investigation was to study the efficacy of markers in differentiating the rice germplasm from diverse source, by comparing clustering based on random SSR markers vs the trait linked/gene based markers. A total of 88 rice germplasm accessions, representing the gamut of rice diversity from land races to elite cultivars were sourced from different parts of India. These were then characterized for morphological and molecular variability.

To study genetic diversity, it is important to know which type of markers can precisely predict the actual or functional variability existing in the germplasm of a crop. In this study, we compared the data generated using random SSR markers vs the SSR markers which have been reported to be gene based or linked to genes related to yield and yield component traits in rice (Zhang et al.2010). It has been considered by earlier researchers that the diversity established by using random markers, which are essentially derived from the functionally inactive or noncoding regions of the genome might not illustrate the ‘functional’ diversity. Therefore, it is important to use markers which could unravel the diversity in functionally related coding regions of the genome to identify the variability.

The results based on the analysis of diversity of germplasm with two different sets of SSRs markers showed that germplasm originating from specific centre clustered together irrespective of the marker sets reflecting the narrow genetic base in the rice breeding programmes at these centres. Therefore, there is an urgent need for broadening the genetic base of breeding programme to improve the per se performance of varieties and the degree of heterosis.

The varieties procured from Bangalore exhibited least genetic diversity since most of these genotypes are RILs derived from the IR 64/Azucena cross. The high level of diversity exhibited by the varieties from Srinagar can be attributed to the inclusion of both indica and japonica varieties in the collection and also because these varieties are well adapted to different ecologies in Jammu and Kashmir, namely intermediate zone (1300–2000 m msl) of Jammu and Kashmir (varieties including SR1, SKAU 5, SKAU 23 and SKAU 27) and varieties suited for high altitude areas (> 2000 m msl) of Jammu and Kashmir (K 332 and K 429).

Both the set of marker data were able to discriminate the landraces from the other varieties. The presence of high genetic similarity among landraces of a common geographic region has been described in earlier studies (Ram et al.2007). However, this rich genetic variability which has accumulated over a period of time is yet to be exploited by plant breeders.

Analysis with both sets of markers classified the germplasm into different groups of landraces, indica varieties, japonica varieties and Basmati groups. These subpopulations seem to have been created during the course of rice domestication (Kovach et al.2007). The indica and Basmati group were found to be the most polymorphic. The Basmati varieties were placed closer to japonica group than indica group. Basmati varieties have conventionally been classified as members of the indica varietal group essentially because of their long and slender grain shape. Previous studies have also, however, reported that Basmati lines are more closely related to japonica than indica group, with the FST value for tropical japonica-aromatic reported to be 0.23 and the FST for indica-aromatic to be 0.39 (Garris et al.2005). The conspicuous differentiation between Basmati and nonBasmati varieties is indicative of divergence and subsequent independent evolution of the former through artificial selection (Nagaraju et al.2002). This kind of classification has been well documented as Glaszmann (1987) classification of Asian rice germplasm. He proposed a six group classification of Asian rice germplasm, with two major groups, group I and group VI corresponding to the indica and japonica group of varieties respectively. Group V represented the Basmati group from Iran, Pakistan, India and Nepal.

MR 219 and MR 220 were the least dissimilar lines. These two varieties are sister lines released in Malaysia. Jaiphulla, a landrace procured from Orissa was distant from the other landraces from the same and different centres. There is little phenotypic or ancestry information about the landraces that can justify this observation. However, previous studies (Singh et al.2010) have pointed out that Jaiphulla is medium in height, early maturing and high yielding variety as opposed to the other landraces which are mostly tall, medium to long duration type with long panicle, having moderate to high yield.

The average genetic diversity at genomic loci as assessed by trait-linked markers was found to be more than that revealed by random markers. This indicates that there exists greater variability at the functional regions of the genome which is rightly established by the functional markers in the selected germplasm. There is an argument that genic SSRs display a low level of polymorphism vis-à-vis genomic SSRs, since SSRs located in the coding regions are under strong selection pressure and therefore accumulate few mutations (Li et al.2004; Varshney et al.2005). This was demonstrated in many crops such as rice (Cho et al.2000), sugarcane (Cordeiro et al.2001), wheat (Gupta et al.2003) and chickpea (Choudhary et al.2009). However, in the present study we have observed slightly higher average PIC values for the trait-linked SSRs (0.48) than the random SSRs (0.41). Apparently, we have used rice microsatellite (RM) markers linked to QTLs affecting yield and yield components (trait-linked markers), but not truly functional gene based markers. Although, they are linked to genomic regions that affect agronomic traits, but not located within the coding region. Therefore, it is unreasonable to expect such markers to fall on conserved regions all the time and to have lesser polymorphism than the random markers. Further, when markers linked to known QTLs are tested in a germplasm with large variation for the respective traits (yield and its components in this study), it is likely that we are bringing in more variable alleles together with respect to the target trait. Therefore, allele richness of any QTL linked marker would have slight advantage over that of a random marker resulting into relatively higher average PIC value for such specific QTL linked markers than random markers.

There was a moderately significant correlation between the phenotypic diversity and kinship as revealed by trait-linked markers. The use of such markers will therefore enable plant breeders to rightly estimate the diversity at heterotic loci. However, the low correlation value confirms role of environment in influencing the phenotype and how classification based on only phenotypic diversity can be ambiguous. Therefore, it can be concluded that both the phenotypic diversity and genetic diversity at functional regions of the genome, should be targeted and utilized in heterotic rice breeding programmes, for predicting hybrid performance.

Notes

Acknowledgements

This study was carried out under the project ‘Marker Assisted Creation of Heterotic Pools and Diversification of Male Sterility and Fertility Restoration System for Hybrid Seed Production in Rice’ funded by the Council of Scientific and Industrial Research (CSIR), Government of India, under The New Millennium Indian Technology Leadership Initiative (NMITLI).

Supplementary material

12041_2013_312_MOESM1_ESM.pdf (514 kb)
(PDF 514 KB)

References

  1. Botstein D., White R. L., Skolnick M. and Davis R. W. 1980 Construction of a genetic linkage map in man using restriction fragment length polymorphisms. Am. J. Hum. Genet. 32, 314–331.PubMedCentralPubMedGoogle Scholar
  2. Cho Y. G., Ishii T., Temnykh S., Chen X., Lipovich L., McCouch S. R. et al. 2000 Diversity of microsatellites derived from genomic libraries and GenBank sequences in rice (Oryza sativa L.). Theor. Appl. Genet. 100, 713–722.CrossRefGoogle Scholar
  3. Choudhary S., Sethy N. K., Shokeen B. and Bhatia S. 2009 Development of chickpea EST-SSR markers and analysis of allelic variation across related species. Theor. Appl. Genet. 118, 591–608.PubMedCrossRefGoogle Scholar
  4. Cordeiro G. M., Casu R., McIntyre C. L., Manners J. M. and Henry R. J. 2001 Microsatellite markers from sugarcane (Saccharum spp.) ESTs cross transferable to erianthus and sorghum. Plant Sci. 160, 1115–1123.PubMedCrossRefGoogle Scholar
  5. Felsenstein J. 1985 Confidence limits on phylogenies: An approach using the bootstrap. Evolution 39, 783–791.CrossRefGoogle Scholar
  6. Ford-Lloyd B. V., Jackson M. T. and Newbury H. J. 1997 Molecular markers and the management of genetic resources in seed genebanks: a case study of rice. In Biotechnology and plant genetic resources: conservation and use (ed. J. A. Callow, B. V. Ford-Lloyd and H. J. Newbury). CAB International, Wallingford, UK.Google Scholar
  7. Garris A. J., Tai T. H., Coburn J., Kresovich S. and McCouch S. 2005 Genetic structure and diversity in Oryza sativa L. Genetics 169, 1631–1638.PubMedCrossRefGoogle Scholar
  8. Glaszmann J. C. 1987 Isozymes and classification of Asian rice varieties. Theor. Appl. Genet. 74, 21–30.PubMedCrossRefGoogle Scholar
  9. Gupta P. K., Rustgi S., Sharma S., Singh R., Kumar N. and Balyan H. S. 2003 Transferable EST-SSR markers for the study of polymorphism and genetic diversity in bread wheat. Mol. Genet. Genomics 270, 315–323.PubMedCrossRefGoogle Scholar
  10. Hammer Ø., Harper D. A. T. and Ryan P. D. 2001 PAST: Paleontological statistics software package for education and data analysis. Palaeontol. Electron. 4, 9.Google Scholar
  11. Kovach M. J., Sweeney M. T. and McCouch S. R. 2007 New insights into the history of rice domestication. Trends Genet. 23, 578–587.PubMedCrossRefGoogle Scholar
  12. Kumar S., Bisht I. S. and Bhatt K. V. 2010 Assessment of genetic diversity among rice (Oryza sativa L.) landrace populations under traditional production using microsatellite (SSR) markers. Electron. J. Plant Breed. 1, 474–483.Google Scholar
  13. Li Y.C., Korol A. B., Fahima T. and Nevo E. 2004 Microsatellites within genes: structure, function, and evolution. Mol. Biol. Evol. 21, 991–1007.PubMedCrossRefGoogle Scholar
  14. McCouch S. R., Chen X. and Cho Y. G. 2002 Sequence divergence of rice microsatellites in Oryza and other plant species. Mol. Genet. Genomics 268, 331–343.PubMedCrossRefGoogle Scholar
  15. Nagaraju J., Kathirvel M., Kumar R. R., Siddiq E. A. and Hasnain S. E. 2002 Genetic analysis of traditional and evolved basmati and non-basmati rice varieties by using fluorescence based ISSR PCR and SSR markers. Proc. Natl. Acad. Sci. USA 99, 5836–5841.PubMedCrossRefGoogle Scholar
  16. Perrier X. and Collet J. P. 2006 DARwin software (http://darwin.cirad.fr/).
  17. Prabhu K. V., Somers D. J., Rakow G. and Gugel R. K. 1998 Molecular markers linked to white rust resistance in mustard Brassica juncea. Theor. Appl. Genet. 97, 865–870.CrossRefGoogle Scholar
  18. Rai M. 1999 Rice germplasm evaluation and enhancement in India: issues, status, options, and future plan of action. Proceedings of the International Symposium on Rice Germplasm Evaluation and Enhancement (ed. J. N. Rutger, J. F. Robinson and R. H. Dilday), pp. 83–91, Arkansas Agricultural Experiment Station, Fayetteville, Arkansas.Google Scholar
  19. Ram S. G., Thiruvengadam V. and Vinod K. K. 2007 Genetic diversity among cultivars, landraces and wild relatives of rice as revealed by microsatellite markers. J. Appl. Genet. 48, 337–345.PubMedCrossRefGoogle Scholar
  20. Rosenberg M. S. and Anderson C. D. 2011 PASSaGE: pattern analysis, spatial statistics and geographic exegesis, version 2. Methods Ecol. Evol. 2, 229–232.CrossRefGoogle Scholar
  21. Saini N., Jain N., Jain S. and Jain R. K. 2004 Assessment of genetic diversity within and among basmati and non-basmati rice varieties using AFLP, ISSR and SSR markers. Euphytica 140, 133–146.CrossRefGoogle Scholar
  22. Singh B., Mishra M. K. and Naik R. K. 2010 Genetic diversity among some traditional aromatic rice (Oryza sativa L) varieties of orissa. Indian J. Agric. Res. 44, 141–145.Google Scholar
  23. Sivaranjani A. K. P., Pandey M. K., Sudharshan I., Kumar G. R., Madhav M. S., Sundaram R. M. et al. 2010 Assessment of genetic diversity among basmati and non-basmati aromatic rices of India using SSR markers. Curr. Sci. 99, 221–226.Google Scholar
  24. Song Z. P., Xu X., Wang B., Chen J. K. and Lu B. R. 2003 Genetic diversity in the northernmost Oryza rufipogon populations estimated by SSR markers. Theor. Appl. Genet. 107, 1492–1499.PubMedCrossRefGoogle Scholar
  25. Sundaram R. M., Naveenkumar B., Biradar S. K., Balachandran S. M., Mishra B., Ahmed M. I. et al. 2008 Identification of informative SSR markers capable of distinguishing hybrid rice parental lines and their utilization in seed purity assessment. Euphytica 163, 215–224.CrossRefGoogle Scholar
  26. Temnykh S., Park D. W., Ayres N., Cartinhour S., Hauck N., Lipovich L. et al. 2000 Mapping and genome organization of microsatellite sequences in rice (Oryza sativa L.). Theor. Appl. Genet. 100, 697–712.CrossRefGoogle Scholar
  27. Vanniarajan C., Vinod K. K. and Pereira A. 2012 Molecular evaluation of genetic diversity and association studies in rice (Oryza sativa L.). J. Genet. 91, 9–19.PubMedCrossRefGoogle Scholar
  28. Varshney R. K., Graner A. and Sorrells M. E. 2005 Genic microsatellite markers in plants: features and applications. Trends Biotechnol. 23, 48–55.PubMedCrossRefGoogle Scholar
  29. Virk P. S., Newbury J. H., Bryan G. J., Jackson M. T. and Ford-Lloyd B. V. 2000 Are mapped or anonymous markers more useful for assessing genetic diversity? Theor. Appl. Genet. 100, 607–613.CrossRefGoogle Scholar
  30. Zhang T., Ni X-L., Jiang K-F., Deng H-F., He Q., Yang Q-H. et al. 2010 Relationship between heterosis and parental genetic distance based on molecular markers for functional genes related to yield traits in rice. Rice Sci. 17, 288–295.CrossRefGoogle Scholar

Copyright information

© Indian Academy of Sciences 2013

Authors and Affiliations

  • SHEEL YADAV
    • 1
    • 5
  • ASHUTOSH SINGH
    • 1
  • M. R. SINGH
    • 2
  • NITIKA GOEL
    • 1
  • K. K. VINOD
    • 3
  • T. MOHAPATRA
    • 4
  • A. K. SINGH
    • 1
  1. 1.Division of GeneticsIndian Agricultural Research InstituteNew DelhiIndia
  2. 2.Post Graduate CollegeGhazipurIndia
  3. 3.Indian Agricultural Research InstituteRice Breeding and Genetics Research CentreAduthuraiIndia
  4. 4.Central Rice Research InstituteCuttackIndia
  5. 5.Division of Genomic ResourcesNational Bureau of Plant Genetic ResourcesNew DelhiIndia

Personalised recommendations