Starch content and cassava mosaic disease genetic diversity with relation to yield in south Indian cassava (Manihot esculenta Crantz) germplasm

179

The Korean Society of Crop Science

J. Crop Sci. Biotech. 2011 (September) 14 (3) : 179 ~ 189

RESEARCH ARTICLE

DOI No. 10.1007/s12892-010-0101-0

Starch Content and Cassava Mosaic Disease Genetic Diversitywith Relation to Yield in South Indian Cassava (Manihot esculenta Crantz) Germplasm

Raghu Duraisamy1, Sakthi Ambothi Rathinasamy1, Senthil Natesan1*, Raveendran Muthurajan1, Jana Jeevan Ramineni1,Nageswari Karuppusamy2, Pugalenthi Lakshmanan3, Mohan Chokkappan4, Karthikeyan Gandhi5

1Centre for Plant Molecular Biology, Tamil Nadu Agricultural University, Coimbatore, India 2Tapioca and Castor Research Station, Tamil Nadu Agricultural University, Yethapur, India3Horticultural Colleges and Research Institute, Tamil Nadu Agricultural University, Coimbatore, India4Central Tuber Crops Research Institute, Trivandrum, Kerala, India5Centre for Plant Protection Studies, Tamil Nadu Agricultural University, Coimbatore, India

Received: November 2, 2010 / Revised: February 8, 2011 / Accepted: August 23, 2011Ⓒ Korean Society of Crop Science and Springer 2011

Abstract

Cassava, family Euphorbiaceae, is the fifth most important staple food crop. The study of variability and diversity present indiverse cassava accessions maintained at the field genebank of TCRS, TNAU is essential to design the breeding program of cassavain TNAU. Hence, a study was carried out to assess the genetic diversity based on morphological traits and Simple Sequence Repeat(SSR) markers, identifying the SSR markers linked to a cassava mosaic disease-resistant QTL region. Genetic diversity among cassa-va germplasm accessions was estimated using 28 morphological traits and three SSR markers. The polymorphic information content(PIC) result gives high information for all the SSR markers. Morphological traits revealed 10 clusters and the SSR markers revealed13 clusters at a similarity coefficient of 2.514 and 0.75, respectively. The SSR marker diversity revealed that the most promisingclusters (II and XII) possessing accessions with yield attributing traits and cluster IX were grouped together due to low yield andCMD susceptibility. Morphological descriptors revealed variance for yield contributing traits. The first three principle componentsaccounted for 10.8, 9.1, and 8.4%, respectively. Correlation studies showed significant correlation among yield and yield-attributingtraits, which in turn influences yield. The morphological traits diversity reveals that cluster VI was the most promising cluster inwhich genotypes ME209 and ME460 possess higher yield and starch content with cassava mosaic disease (CMD) resistance. Most ofthe accessions grouped under cluster X belong to selection from International Centre for Tropical Agriculture (CIAT) collections andthe cultivars under clusters I and II were true cassava seed (TCS) selections of India. The CMD-linked SSR marker which is report-ed in the previous studies have generated eight alleles and each were regressed with CMD resistance and resulted in non-significantlinkage to CMD resistance.

Key words: Cassava germplasm, genetic diversity, morphological markers, SSR markers

Cassava (Manihot esculenta Crantz), belongs to the familyEuphorbiaceae, with a diploid genome (2n = 36). Of the 98species that belong to the genus Manihot, cassava is the onlyspecies that is widely cultivated for food production. It is a

major source of low-cost carbohydrates for populations in thehumid tropics, requiring at least 8 - 9 months of warm weatherto produce a crop. All species of the genus Manihot are native tothe New World tropics. Cultivated and wild accessions wereintroduced to the Old World tropics in the 16th century. Cassavawas introduced in India by Portuguese in Malabar region duringthe 17th century as a food crop gradually changing its role as an

Introduction

Senthil Natesan ( )E-mail : [email protected]: +919842232057

Cassava Genetic Diversity180

industrial raw material. Cassava was popularized in Kerala bythe king of Travancore state, Sri Vishakham Thirunal. He intro-duced high-yielding varieties from Malaya and other places. TheCentral Tuber Crops Research Institute (CTCRI), Trivandrumreleased 13 varieties with improved yield and high starch con-tent and Tamil Nadu Agricultural University (TNAU) has alsoreleased two varieties viz. CO1 and CO2 for commercial culti-vation in Tamil Nadu.

Conservation of genetic resources of crop plants, todayregarded as an urgent priority, requires characterization of thegenetic diversity of domesticated plants and understanding itsorigin and evolution under domestication (Miller et al. 1995).The largest germplasm collection for cassava is housed at CIATCali, Colombia. In India, especially in Tamil Nadu, 367 cassavagermplasm accessions are being maintained at Tapioca andCastor Research Station (TCRS) which include CIAT andCTCRI collections apart from the native collection in Yethapur(Raghu et al. 2007).

A prerequisite for any program of genetic improvement ofcassava is knowledge of the extent of genetic variation presentbetween cultivars and the genetic distance between cassava andclosely related species with which hybrids could be produced.For assessment of genetic diversity and genetic improvement,germplasm collections are vitally important. Genetic divergenceamong the genotypes plays an important role in selection of par-ents having wider variability for different traits which can beused in hybridization programs (Nayak et al. 2004). The crossesbetween parents with maximum genetic divergence are general-ly most responsive for genetic improvement (Arunachalam1981). The cassava germplasm accessions were collected fromdifferent regions of India through explorations and maintained atTCRS, TNAU, and Yethapur, and used for various cropimprovement programs. The clones of different cassava acces-sions were maintained through clonal propagation. These acces-sions were raised every year and outstanding varieties wereselected which can be used for hybridization programs. The

present study is aimed at understanding the genetic diversitybased on morphological and SSR makers and at identifying theSSR markers linked to CMD resistance using previously report-ed SSR markers linked to a CMD resistant QTL region.

Materials and Methods

Plant materialThe subset of 255 diverse cassava germplasm accessions

(Table 1: supp.) were selected for evaluation of genetic diversityfrom 367 germplasm lines maintained at TCRS. The plants weregrown under field conditions in south-west monsoon season(July 2007). Each accession was raised in a single row of 7.5meters in length by adopting a spacing of 90 x 75 cm. All therecommended agronomic packages of practices were adoptedduring the entire crop period and the observations on variousmorphological traits were recorded. The 28 morphological traits(10 quantitative and 18 qualitative) were recorded in three plantsper accession based on the descriptors (genetic diversity of culti-vated tropical plants). The traits recorded are given in Table 1.

Screening germplasm for CMD resistanceThe level of resistance to CMD across 255 cassava germplas-

m accessions was done after 6 months set planting based on thescoring system adopted by Hahn et al. (1980b) under naturalinfected field conditions. The level of resistance was determinedbased on 1-5 scores (Suppl. Table 2). The damage score estab-lished is as per the severity of the symptoms developed (Fig. 1and Suppl. Table 2).

Simple sequence repeats (SSR) diversityThree CMD-linked primer pairs namely SSRY324, SSRY28,

and SSRY2 were used to carry out the SSR marker survey in theselected 103 diverse cassava accessions (Suppl. Table 3). The

Table 1. Descriptive statistics for morphological traits recorded in morphological diversity among Cassava accessions

PS BC EE AI SL CYL SLL PP CUSL CLN TG CUNL NTYSC CP CMD ISC PC ST OSC PH NLL LP NB Srarch NL Yield/Plant

TL

Mean

Standard error

Standard devitation

Sample variance

Coefficientof variance

1.31

0.03

0.55

0.31

42.47

1.65

0.03

0.48

0.23

28.93

1.17

0.02

0.38

0.14

32.1

1.56

0.03

0.58

0.34

37.33

1.56

0.03

0.58

0.34

37.33

1.76

0.07

1.07

1.15

60.92

2.65

0.06

0.95

0.91

35.89

1.1

0.02

0.3

0.09

27.13

2.03

0.06

0.96

0.92

47.22

1.01

0.01

0.13

0.02

12.43

1.38

0.06

0.92

0.85

66.67

2.15

0.07

1.14

1.29

52.95

2.67

0.05

0.78

0.61

29.26

3.02

0.06

0.95

0.89

31.3

1.24

0.04

0.59

0.35

47.61

1.19

0.04

0.64

0.41

53.58

3.76

0.05

0.73

0.53

19.34

1.99

0.05

0.83

0.69

41.74

1.58

0.02

0.33

0.11

20.78

6.1

0.05

0.75

0.57

12.34

15.43

0.2

3.18

10.09

20.58

1.88

0.05

0.76

0.58

40.62

20.14

0.21

3.42

11.66

16.96

19.31

0.36

5.69

32.33

29.44

17.79

0.26

4.13

17.05

23.21

7.41

0.31

4.99

24.89

67.3

1.92

0.02

0.27

0.073

14.02

1.47

0.06

0.9

0.8

60.85

PS: Shape of the plant; BC: Colour of the young bud; EE: Emergence of eye; AI: Alignment of internodes; SL: Shape of the leaf; CYL: Colour of the young leaf; SLL: Shape of theleaf lobes; PP: Petiole pubescence; CUSL: Colour of the upper surface of lamina; CUNL: Colour of the under surface of lamina; CLN: Colour of the lamina nerve; YSC: Colour ofthe young stem; CP: Colour of the petiole; CMD: Cassava mosaic disease; ISC: Inner skin colour of the tuber; PC: Pulp colour; ST: Shape of the tuber; OSC: Outer skin colour ofthe tuber; PH: Height of the plant; NLL: Number of leaf lobes; LP: Length of the tube; NB: Number of branches; TL: Length of the tuber; TG: Girth of the tuber; NT: Number oftubers per plant; NL: Neck length of the tuber.

JCSB 2011 (September) 14 (3) : 179 ~ 189 181

PCR amplification reactions were performed in a 20 µL volumeusing 50 ng of DNA containing 0.2 µM of each forward- andreverse-primers, 20 µM of dNTPs, 2 µL (1X) Taq polymerasebuffer, 1.5 mM MgCl2, and 0.3 U Taq DNA polymerase. TheSSR profiling was carried out using PTC100 (M/s. MJ Research,Inc.) Thermal Cycler Program. The main program was: initialdenaturation of 5 min. at 94°C then 30 cycles of 1 min. at 94°C,2 min. at 58°C, 2 min. at 72°C, and a final extension of 5 min. at72°C. PCR products were run on 4% agarose gel electrophoresisafter denaturation.

Descriptive statistics, correlation, PCA, NTSYS, andmarker-trait association studiesDescriptive statistics analysis The 28 morphological traits wererecorded in 255 germplasm accessions. The mean values ofrecorded 28 morphological traits were analysed for descriptivestatistics using MINITAB15 program.

Principle component analysis (PCA) Factor analysis was per-formed among 28 traits to know which trait is contributing tomaximum variability by unrotated factor loading. The number oftraits were reduced by factor analysis and further subjected toPCA using the MINITAB15 program. The PCA will dissect thetotal variability into different components like PCA I, II, and III.The first PCA component always possesses a higher per cent ofvariation in the total variation than the last component. So in this

study, the first three PCA components were taken for trait vari-ability studies since it represents maximum variability of acces-sions.

Pearson correlation analysis The traits showing variation aresubjected to Pearson correlation analysis (yield contributingquantitative traits) using SPSS 10.0 statistical software program.Polymorphism information contents (PIC) were calculatedaccording to Anderson et al. (1993) using the following simpli-fied formula:

PIC = 1-∑pi2

Where pi is the frequency of the ith allele for marker across allalleles for the locus (Nei 1973).

Diversity analysis SSR maker alleles were scored as 0/1 forabsence/presence of alleles, respectively. Similarity coefficientmatrices were calculated using the Jaccard similarity algorithm(Jaccard 1908) for SSR and morphological markers. Dendrogra-ms were constructed by SIMQUAL and SIMINT programme ofNTSYS-PC, version 2.02i (Rohlf 1998) for SSR and morpholog-ical markers, respectively, using the UPGMA method(Unweighted Pair-Group Method with arithmetical algorithmsAverages; Sneath and Sokal 1973; Sokal 1995).

Regression analysis In order to detect the marker-trait associa-tion, both simple linear regression and multiple regressionanalysis were applied. The CMD resistance served as dependentvariables and the SSR marker allele as independent variable. Forsimple linear regression, data on individual marker alleles wereregressed with CMD resistance score using Excel 2007 program.Multiple regression analysis was performed using a computer-aided software MINITAB version 15.

Results

The genetic diversity of different cassava accessions wereseparated out using morphological and SSR markers (Tables 1and Suppl. Table 3). The nature of variability was also analysedusing descriptive statistics and Principle component analysis(PCA) using total variability present among germplasm acces-sions (Suppl. Table 1). Correlation study was done to explain theindividual trait correlation among morphological traits. Themarker-trait (CMD resistance) association study also performedto unveil the nature of association between the markers linked toa CMD resistant QTL and the CMD phenotype. The results aredescribed as below.

Variability analysisThe cassava genetic improvement programme needs extent

knowledge of genetic variations in the available Manihotspecies. The observations of 28 morphological traits were sub-jected to descriptive statistics analysis viz. standard error, stan-dard deviation, sample variance, and coefficient of variation andthe values are presented in Table 1. Considerable variationsFig. 1. Severity of the disease and the respective damage score.

were observed for TL followed by NT and TG. The coefficientof variation was found to be the maximum for NT (67.3) andminimum for NLL (12.34). The naturally occurring variation inCMD (Fig. 1), yield per plant, and starch content indicates thepossibility for improving yield along with the available geneticstocks. Variations found in qualitative traits are more useful indeveloping varietal identification descriptors whereas accessionspossessing variable quantitative traits are used in varietalimprovement programs. Frequency distribution of yield, starchcontent, and CMD are given in the Figs. 2A, B, and C.

Contribution of traits towards varianceThe factor analysis was carried out for all the 28 morphologi-

cal traits to identify the combination of traits contributing formaximum variability. The first factor had a contributing unrotat-ed factor loading of 9.4%, which was contributed by the traits,color of the young leaf (CYL), color of lamina nerve (CLN),color of the petiole (CP), pulp color (PC), TG, and NT. The sec-ond factor had a maximum factor loading of 8.2% contributedby the traits bud color (BC), height of the plant (PH), length ofthe petiole (LP) and CP, shape of the leaf (SL), CYL, and shapeof the leaf lobe (SLL), color of the under surface of lamina(CUNL) and yield per plant. The third factor had a contributingfactor loading of 7.6%. The variability is contributed by thetraits TL, CMD, number of branches (NB), PC, alignment ofinternodes (AI), BC, and plant shape (PS). Among these 28 vari-ables, petiole color was found to be of maximum variation(60.8%) and neck length (NL) of the tuber had the least variation(2.1%), whereas BC and NT had contributing variation of 57

and 49.5%, respectively, in total variation (25.1%). For principalcomponent analysis, 24 traits which had more than 30% varia-tion were selected from the first three factors.

Among 28 morphological traits, 24 which contributed consid-erably to the total variation were selected based on unrotatedfactor loading (factor analysis) in order to reduce the number oftraits to perform the principal component analysis (Fig. 2). Thefirst principal component accounted for 10.8% variation withhigh contributing factor loading from NT, PC, CP, PH, andCLN. The second principal component accounted for 9.1% vari-ation and high factor loadings contributed by BC, AI, SLL, peti-ole pubescence (PP), and CP. The third principal componentaccounted for 8.4% variation with a very high contributed factorloading by TL, yield per plant, TG, PH, PS, NB and CMD. PH,TL, and CP are the traits which showed the highest variationamong the genotypes. The total variation contributed by thesetraits was 28.2%. The traits which contribute to maximum vari-ability can be identified and the accessions possessing thesevariable traits can be chosen for further breeding programs.

Correlation between yields associated traitsSelection in breeding programs depends on the magnitude ofvariability in the population and the relationship between thekey morphological traits. The correlation analysis was done forindirect selection of morphological traits which are related tocontribution of the yield. The estimate of correlation coefficientbetween six different quantitative traits which are associatedwith yield and their relationship are presented in Table 2. TheNT (0.479) possesses positive significant correlation with TG,which in turn influences yield and negative correlation is signifi-cantly associated with TL (-0.172) and starch content (-0.182) ofthe tuber at 1% level. The TL (0.133) possesses significant cor-relation with the yield directly at 5% level. The yield can beimproved by selecting any one of these traits. The positivelyassociated traits can be taken into consideration while selectingparents for hybridization work.

Phenogram pattern of cassava accessions based onmorphological traits

The similarity index for the recorded morphological traitsamong the cassava genotypes was worked out by calculating theJaccard’s coefficient. The similarity coefficients were based on28 morphological markers ranging from 0.6 to 3.97. Among the

Cassava Genetic Diversity182

Fig. 2 Number and distribution of traits among 255 Cassava (Manihot esculentaCrantz) accessions from the subset of germplasm. Black bars indicate number ofaccessions unique to cassava. (A) Distribution of yield (kg) per plant; (B) Distributionof starch content of the tuber; (C) Distribution of Cassava Mosaic Disease score.

StarchTLTGNTNL

Yield per Plant

1 .0000 .1130 .040-0.182**

0 .055-0.090

1 .0000 .060-0 .172**

0 .0110 .133*

1 .0000 .479**

-0.0230 .116

1 .000-0 .1160 .084

1.0000.043 1.000

Table 2. Pearson correlation for yield-associated traits among Cassavaaccessions.

Starch TL TG NT NLYield per

Plant

**. Correlation is significant at the 0.01 level.*. Correlation is significant at the 0.05 level.

JCSB 2011 (September) 14 (3) : 179 ~ 189 183

255 accessions studied, the highest similarity index (2.062) wasrecorded between the genotypes Me.729 and Me.742. The low-est similarity index (0.622) was recorded between the genotypesMe.337 & Me657 and Me256 & 259. The similarity valuesobtained for each pairwise comparison of morphological mark-ers among the 255 cassava accessions were used to constructdendrogram (Fig. 4) based on hierarchial clustering. The cassavaaccessions were grouped into 10 clusters at a similarity index of2.514.

The cluster distance was estimated among the clusters basedon the similarity coefficient. Clusters IV and V were found to bea maximum distance of 1.046 and a minimum distance of 0.011were recorded in cluster III on clusters IX and X. Whereas clus-ters IX and X had zero distance. The cluster size varied from 2to 138. Cluster X is the largest cluster comprising of 138 acces-sions followed by cluster V, cluster IX, cluster VIII, cluster I,and cluster III possessing 62, 24, 12, 7, and 4 genotypes respec-tively. Cluster II, cluster IV, cluster VI, and cluster VII possess-es 2 accessions each. Cluster VI was most promising cluster inwhich genotypes ME 209 and ME 460 possess higher percent ofstarch content (30 and 27) and yield per plant (4 and 6.33 kg)with CMD resistance. Accessions, namely ME181, ME187,ME189, ME193, ME199, ME211, and ME253 also had higherstarch (> 27%) content. However, these accessions weregrouped into cluster X due to lesser yield with high susceptibili-ty to CMD. Cluster II consisted of two accessions, ME729 andME742, each possessing higher TL (43) with less starch content(< 17%). Whereas ME205 and ME315 possessed higher TG (>23.7 cm) with medium TL (> 26.7) grouped into a separate clus-ter (cluster VII). Cluster III containing ME697, ME703, ME706,and ME712 were found be higher in NT (> 12) per plant.ME299 and ME705 accessions come under the cluster IV due tothe unique nature showing moderate per cent of starch (20 - 23)as well as TG (24 - 28 cm) and less NT (3 - 6). Cluster V whichincluded ME337, ME464, ME544, ME600, ME656, ME657,ME659, ME682, and ME690, possessed thin tubers (< 11 cm),so these were grouped into a separate cluster. Most of the acces-sions grouped under cluster X belong to selection from CIATcollections and the cultivars under clusters I and II were TCSselections.

Table 3. Genetic distance among different clusters based on similaritycoefficient estimated from SSR marker diversity.

Cluster I II III IV V VI VII VIII X IX XI XII XIII

IIIIIIIVVVIVIIVIIIIXXXIXIIXIII

0.000

0.134

0.134

0.084

0.084

0.072

0.149

0.149

0.088

0.024

0.143

0.143

0.034

0.000

0.000

0.050

0.050

0.063

0.005

0.005

0.222

0.159

0.009

0.009

0.100

0.000

0.050

0.050

0.063

0.005

0.005

0.222

0.159

0.009

0.009

0.100

0.000

0.000

0.014

0.045

0.045

0.172

0.109

0.059

0.059

0.050

0.000

0.014

0.045

0.045

0.172

0.109

0.059

0.059

0.050

0.000

0.057

0.057

0.159

0.095

0.072

0.072

0.034

0.000

0.000

0.217

0.154

0.014

0.014

0.095

0.000

0.217

0.154

0.014

0.014

0.095

0.000

0.063

0.230

0.230

0.122

0.000

0.054

0.054

0.059

0.000

0.000

0.109

0.000

0.109 0.000

Fig. 4 Phenogram of Cassava (Manihot esculenta Crantz) accessions based on mor-phological traits. The dendrogram tree obtained from Unweighted Pair-Group(UPGMA) cluster method of Nei's genetic distances option in the NTSYS-PC program.

Fig. 3. Factor loading plot for principle component analysis among twenty-four mor-phological traits selected from factor analysis based on variability.

III

IIIIV

V

VI VII

VIII

X

-0.3-0.4

-0.3

-0.2

-0.1

0.0

0.1

0.2

0.3

0.4

-0.2 -0.1 0.0First Component

Seco

ndCo

mpo

nent

0.1 0.2 0.3 0.4

IX

Phenogram pattern of cassava accessions based on SSRmarkers

The total number of polymorphic alleles generated from SSRprimers ranging from 138 bp for the primer pair SSRY2 to 189bp for the primer pair SSRY324. The PIC for the SSR primerswas generally high for SSRY28 primer pair and SSRY324(Suppl. Table 3). Whereas primer pair SSRY2 showed less PIC(0.46). The SSR analysis was able to separate cassava acces-sions on different groups based on their genetic distances. TheJaccard’s similarity coefficient for the SSR alleles were calculat-ed and phylogenetic tree was constructed using UPMA,NTSYS-PC (Fig. 5). The 103 cassava accessions were groupedinto 13 clusters at coefficient of 0.75 (Fig. 5). The cluster dis-tance was estimated among the different clusters based on thesimilarity coefficient which ranges from 0.005 between clustersII and III, and VII and VIII to 0.230 between clusters IX and XI& XII (Table 3). The cluster size varied from 1 to 23 and clusterXIII was the largest cluster comprising of 23 accessions fol-lowed by cluster VIII, VI&XI, III, XII, II, IV, VII V, IX, X, andI possessing 13, 11, 8, 7, 5, 5, and 1 genotypes, respectively.Clusters II and XII were most promising cluster possessinghigher TG (> 23.7 cm), TL (> 30 cm), and LP (> 20 cm).Accessions namely ME261, ME434, and ME641 were groupedinto separate cluster (IX) due to their cylindrical form of ST,highly susceptible to CMD, and low yield. Clusters VII and XIIIhad higher yield accessions and clusters V and III had low yield,though these clusters varied in TG, and shape of the tuber (ST).Some of the clusters had unique traits such as cluster I withmore NT and LP, cluster IV with pedunculate type of tuberattachment, cluster VI with severe CMD, cluster X with project-ing emergence of eye (EE), cluster VIII with moderate level ofyield, and cluster XI with white inner skin color of the tuber(ISC) and presence of petiole pubescence.

Association of SSR alleles with CMDThe CMD-linked SSR marker which is reported in the previ-

ous studies viz. SSRY324, SSRY28, and SSRY2 generated eightalleles. Each of the CMD-linked SSR maker alleles were

regressed with CMD resistance and resulted in weak R2 valueranged from 0.001 to 0.017 which indicates non-significant link-age of SSR marker alleles to CMD resistance.

Discussion

Variability refers to the degree of difference among the indi-viduals of a population. It is due to difference either in thegenetic constitution of the individual plants or in the environ-ment in which they are grown. The success of genetic improve-ment in any character depends on the nature of variability pres-ent in the gene pool for that character. Hence, assessment ofexisting variability for any character in the gene pool of a cropspecies is of utmost importance to a plant breeder for starting ajudicious plant breeding program. In this study, the morphologi-cal variation did exist among the 255 cassava genotypes withrespect to the 28 morphological traits recorded. Fukuda andAlves (1987) analyzed variation at 21 agronomic traits in the 94cassava accessions. The extent of variability was found to bemaximum NT (67.3) followed by CLN (66.67), BC (60.92), andyield per plant (60.85). The estimation of descriptive statistics of28 different morphological traits studied revealed the existenceof high level of morphological diversity among the cassavagenotypes, providing scope for improvement through hybridiza-tion and selection. Raghu et al. (2007) also found a significantlevel of variation among the quantitative and qualitative traits incassava germplasm. Similarly, Carvalho and Schaal (2001)reported a high degree of genetic variability among 94 cassavaaccessions of Brazilian origin. Pillai et al. (2001) also observedhigh variability for yield per plant and starch content while eval-uating the CMD symptom-free genetic stocks of cassavagermplasm.

Factor analysis was performed in order to reduce a large setof morphological traits (28) to a more meaningful smaller set oftraits (24) and to identify the trait contributing for maximumvariability. Factor analysis provides an exact picture of variabili-ty contributed by each trait. On the basis of factor loadings ofthe 28 morphological traits, the three factors contributed most tovariability were selected for principal component analysis. Thefirst three factors contributed around 25.1% of the total varianceobserved. Hence, 24 traits viz. PS, BC, EE, AI, CYL, SLL, PP,color of the upper surface of lamina (CUSL), CLN, young stemcolor (YSC), CP, PH, LP, NB, CMD, ISC, ST, outer skin color(OSC) of the tuber, PC, starch content, TL, TG, NT, and yieldper plant were selected for principal component analyses.Similarly, Raghu et al. (2007) also recorded a higher level ofvariation at individual traits for 58 cassava accessions.

A set of 24 diverse morphological traits selected based onfactor analysis was used to group the accessions based on princi-pal component analysis. Distribution among the different acces-sions showed that the first two PCs had Eigen values of 2.581and 2.881, cumulatively accounting for 19.1% of the total varia-tion. The results are in consistent with Carvalho and Schaal(2001) and Raghu et al. (2007), who reported that the quantita-tive traits plant height and root weight were the most important

Cassava Genetic Diversity184

Fig. 5. Phenogram of Cassava (Manihot esculenta Crantz) accessions based onmolecular data. The dendrogram tree obtained from Unweighted Pair-Group(UPGMA) cluster method of Nei’s genetic distances option in the NTSYS-PC program.

JCSB 2011 (September) 14 (3) : 179 ~ 189 185

for discriminating cassava cultivars. The traits which contributefor maximum variability can be identified and the accessionspossessing those variable traits can be chosen for further breed-ing programs.

Selection in breeding programs depends on the magnitude ofvariability in the population and the relationship between keymorphological traits. Correlation provides information on thenature and extent of relationship among the various morphologi-cal traits. The estimates of correlation coefficient between thedifferent traits indicate the extent of direction of association.Correlation between traits are important for three reasons: (1)the genetic causes of correlation through the phenotypic actionof genes, (2) the changes brought about by selection, and (3)natural selection where the relation between quantitative traitsand fitness is the primary agent that determines the genetic prop-erties of that character in a natural population (Falconer 1981).Single-plant yield is a complex character influenced by a largenumber of other component traits. A knowledge of the associa-tion between yield and its component traits helps in improvingthe efficiency of selection. In the present investigation, correla-tion coefficients were worked out between the six yield-con-tributing quantitative traits. Among quantitative traits, tuberlength was found to be significantly associated with yield. Fromthese results, it is evident that these traits are associated withtuber yield. The tuber girth had significant positive correlationwith number of tubers per plant. It indicates that the selection forany of these yield-attributing traits will lead to improvement ofthe other related traits and thereby finally enhancing the tuberyield, hence selection based on TG, NT, and TL will help us toimprove the yield. The NT per plant showed a significantly neg-ative correlation with starch content and TL. This result is inaccordance with Ntuiv et al. (2006), where NT is significantlycorrelated with TG.

The number of tubers was significantly and positively corre-lated with TG, which in turn influenced yield and was signifi-cantly and negatively associated with TL and starch content ofthe tuber, but TL was significantly correlated with yield whichindicates that TL is an important yield component.

Morphological diversity among cassava accessionsIn this study, an attempt was made to determine the extent of

diversity among 255 cassava accessions using the morphologicaltraits. These easily observable, morphological traits are usefulfor preliminary evaluation, because they offer a fast and usefulapproach for assessing the extent of diversity. After similarityindex analysis and hierarchial clustering based on morphologicalobservations, 255 cassava accessions were found to be groupedinto 10 clusters at a similarity index of 2.514. The cluster dis-tance was estimated among the clusters based on the similaritycoefficient. A similar study was carried out in cassava usingmicrosatellites by Goldstein et al. 1995. Among the differentclusters, the cluster size and their distance varied from 2 to 138and 0 to 1.046, respectively. Cluster VI was found to compriseof 2 accessions namely, ME 209 and ME 460 possess higherstarch content (30 and 27) and yield per plant (4 and 6.33 kg)with Cassava mosaic disease resistance. The accessions (given

in results) which had low yield and CMD susceptible werefound to be in cluster X, though it had higher starch (> 27%)content. Manihot accessions which had higher tuber length,number of tubers per plant, and tuber girth were grouped intoclusters II, III, and VII, respectively, and lower tuber number perplant and tuber girth were found to be in clusters IV and V,respectively (data given in results). Most of the accessionsgrouped under cluster X belong to selections from CIAT collec-tions and the cultivars under clusters I and II were true cassavaseed (TCS) selections. The morphological diversity data helpedto differentiate CIAT accessions in a separate cluster (cluster X)and TCS lines in two different groups. The clustering pattern ofCIAT and TCS collections into separate clusters may be due tothe fact that they were domesticated in the same ecologicalzones with narrow genetic base (Moyib et al. 2007; Tonukari etal. 1997).

The accessions Me205 and Me315 (cluster VII), havingunique traits like the high starch content, tuber length and tubergirth grouped in separate cluster. Me299 and Me705 (clusterIV), which is has unique traits such as high petiole length andtuber girth is present in few cassava varieties and hence it isgrouped in a different cluster. Similarly, Me 697, Me703,Me706, and Me712 (cluster III) have the character of high tuberneck length and hence were placed in a different cluster. Similarstudies have been carried out previously for the estimation ofgenetic diversity in cassava. Raghu et al. (2007) reported a highdegree of genetic variability among 58 accessions of southIndian cassava germplasm and they were grouped into 6 distinctclusters based on morphological traits. Similarly, Carvalho andSchaal (2001) also reported a high degree of genetic variabilityamong 94 cassava accessions of Brazilian origin. These acces-sions were grouped into 22 distinct clusters based on morpho-logical traits. Fregene et al. (2000) analyzed the genetic relation-ship between African and Latin American accessions using themorphological traits and AFLP markers although some overlapwas found between them with regard to cassava mosaic disease.

SSR allele diversity among cassava accessionsMolecular diversity was assessed by three CMD-resistance

linked SSR markers and the number of alleles produced by themranged from 2 to 3 with the 100% polymorphism level.Similarly, Chavarriaga-Aguirre et al. 1999 also used microsatel-lite markers in Central American cassava accessions and thenumbers of alleles generated by the SSR markers were higher(minimum six alleles). To characterize the capacity of eachmarker to estimate the level of polymorphic loci, the PIC wascalculated for each marker. The results showed that only twomarkers, namely SSRY 324 and SSRY28, were informativeamong the three tested markers. Therefore, these markers can beused in future studies to investigate genetic diversity and phylo-genetic relationships. Similar studies were also conducted insouth Indian cassava germplasm accessions by Raghu et al.(2007). SSR marker profiles resulted in 13 clusters precisely at75% Jaccard’s similarity coefficient. Most of the high-yieldingaccessions namely, Me293, Me320, Me366, Me598, Me650,and Me710, and accessions (Me401, Me438, Me440, Me491,

Me595, and Me715) possessing yield-attributing traits such asTG, NT, and TL were grouped together with a similarity coeffi-cient of 73.17 and 73.68%, respectively. Carvalho and Schaal(2001) and Raghu et al. (2007) reported a similar finding that thePCR-based markers indicate a strong grouping of varieties relat-ed to the region of cultivation and traits in Brazil and India,respectively. Low starch containing accessions comes undercluster V with a coefficient of 69.66%. Most of the cassavaaccessions were classified in cluster XIII including high-yield-ing accession Me209 at 80.85 coefficient level. Elias et al.(2004) studied the genetic diversity in 38 accessions of cassavacollected from South America and detected a higher degree ofvariability in the land races than in the improved lines. The spe-cific allele linked to sweetness of the tuber was also reported.

Association of SSR marker alleles linked with CMD-resistant QTL

The molecular marker linked with QTL/gene for traits ofinterest are being routinely developed in several crops usingmapping populations such as F2, RILS, DH populations, etc.However, non-availability of mapping populations, absence oftight linkage map, and time needed to develop such populationsare major limitations in the identification of molecular markersfor specific traits. To overcome these limitations, molecularmarkers for traits of interest have been identified usinggermplasm collections (Breseghello and Sorrells 2005;Maccaferri et al. 2005; Pakniyat 1997; Skot et al. 2005; Virk etal. 1996). In the present study, the association analysis has beenperformed for the CMD-resistant QTL region using three SSRmarkers following both the simple linear regression and multipleregressions. Among the SSR markers employed in the presentstudy, none were associated with the CMD resistance due totheir weak R2 value. Hence, a more detailed study with SSRmarkers is required.

Conclusions

Gene tic diversity in crop plants can be measured using vari-ous tools including morphological and molecular markers. TheseSSR markers are a useful tool for a preliminary evaluationbecause they offer a fast and useful means for assessing thegenetic diversity among cassava genotypes. The tuber yield-associated traits viz. TL, TG, and NT are highly informative forselecting high yielding accessions. The marker SSRY 324 dif-ferentiated the CMD resistant, susceptibility and high starchcontent genotypes. This Study resulted in the superior genotypesviz., ME 209, ME 460, ME181, ME187, ME189, ME193,ME199, ME211, and ME253 were identified for high starchcontent and tuber yield, which will help for designing better cas-sava breeding programs in the future

Acknowledgements

This study was financially supported by a Rajiv GandhiNational Fellowship Grant (No: F. 14-2 (SC)/2008 (SA-III) Dt:14.09.2009) from the University Grant Commission (UGC),New Delhi, India to Mr D. Raghu, PhD Scholar, DPMB&B,CPMB, TNAU. The financial support from DBT underProgramme support for Research and Development inAgricultural Biotechnology (PSR&DAB) to Dr. N. Senthil andDr. M. Raveendran is kindly acknowledged. The Cassavagermplasm field maintenance support by Dr. K. Shanmugam,Professor and Head and Mr. Perumal, TCRS, Yethapur deserveappreciation and acknowledgment

References

Arunachalam V. 1981. Genetic divergence in plant breeding. Indian J. Genet. 41: 226-236

Breseghello F, Sorrells ME. 2005. Association mapping of kernel size and milling quality in wheat (Triticum aestivum L.) cultivars. Genetics 105: 445-486

Carvalho LCB, Schaal BA. 2001. Assessing genetic diversity in the cassava (Manihot esculenta Crantz) germplasm collection in Brazil using PCR-based markers. Euphytica 120: 133-142

Elias M, Muhlen GS, Mckey D, Roa AC, Tohme J. 2004. Genetic diversity of traditional South American landraces of cassava (Manihot esculenta Crantz): an analysis using microsatellites. Econ. Bot. 58: 242-256

Falconer DS. 1981. Introduction to quantitative genetic. Second Edition, Longman, New York

Fregene M, Bernal A, Duque M, Dixon A, Tohme J. 2000. AFLP analysis of African cassava (Manihot esculenta Crantz) germplasm resistant to the cassava mosaic disease (CMD). Theor. Appl. Genet. 100: 678-685

Fukuda WMG, Alves AAC. 1987. Banco ativo de germoplasma de mandioca do centro Nacional de Mandioca e Fruticultura. Rev. Bras Mandioca 6: 65-97

Goldstein AT, Linares LL, Sforza C, Feldman MW. 1995. An evaluation of genetic distances for use with microsatellite loci. Genetics 139: 463-471

Hahn SK, Terry ER, Leuschner K. 1980. Breeding cassava for resistance to cassava mosaic disease. Euphytica 29: 673-683

Maccaferri M, Sanguineti MC, Noli E, Tuberosa R. 2005. Population structure and long-range linkage diseqilibrium in a durum wheat elite collection. Mol. Breed. 15: 271-289

Miller K, Allegretti MH, Johnson N, Johnson B. 1995. Measures for conservation of biodiversity and sustainable use of its components. In VH Heywood, ed, Global Biodiversity Assessment, United Nations Environment Programme, Cambridge University Press, UK

Moyib OK, Odunola OA, Dixon AGO. 2007. SSR markers reveal genetic variation between improved cassava cultivars and landraces within a collection of Nigerian cassava germplasm. Afr. J. Biotechnol. 6: 2666-2674

Nayak AR, Chaudhaury D, Reddy JN. 2004. Genetic divergence in scented rice. Oryza 41: 79-82

Nei M. 1973. Analysis of gene diversity in subdivided popula

Cassava Genetic Diversity186

JCSB 2011 (September) 14 (3) : 179 ~ 189 187

tion. Proc. Natl. Acad. Sci. USA. 70: 3321-3323Ntui VO, Uyoh EA, Affangideh U, Udensi U, Egbonyi JP. 200

6. Correlation and genetic variability in cassava (Manihot esculenta Crantz). Int. J. Food Agric. Environ. 4: 147-150

Pakniyat H. 1997. AFLP variation in wild barley (Hordeum spontaneum C. Koch) with reference to salt tolerance and associated ecogeography. Genome 40: 332-344

Pillai SV, Geethakrishnan Nair P, Nair RR. 2001. Evaluation of CMD symptom free genetic stocks from cassava germplasm for quality attributes. J. Root Crops 27: 38-40

Raghu D, Senthil N, Saraswathi T, Raveendran M, Gnanam R et al. 2007. Morphological and simple sequence repeats (SSR) based DNA fingerprinting of south Indian cassava germplasm. Int. J. Integr. Biol. 1: 141-148

Rohlf FJ. 1998. NTSYS–PC, Numerical Taxonomy and Multivariate Analysis System, version 2.02i. Exter software, Setauket, NY

Skot L, Humphreys MO, Armstead I, Heywood S, Skot KP et al. 2005. An association mapping approach to identify flowering time genes in natural populations of Lolium perenne (L.). Mol. Breed. 15: 233-245

Sneath PHA, Sokal RR. 1973. Numerical Taxonomy. Freeman, San Francisco, Theor. Appl. Genet. 93: 613-617

Sokal RR, Rohlf J. 1995. Biometry: The Principle and Practice of Statistics in Biological Research, 3rd ed, WH Freeman, New York

Tonukari NJ, Thottapilly G, Mignouna HD. 1997. Genetic polymorphism of cassava within the Republic of Benin detected with RAPD Markers. Afr. J. Crop Sci. 5: 219-228

Virk PS, Newbury HJ, Jackson MT, Ford-Lloyd BV. 1996. The identification of duplicated accessions with in a rice germplasm collection using RAPD analysis. Theor. Appl. Genet. 90: 1049-105

123456789

101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354

Suppl. Table 1. Germplasm accessions of Cassava (Manihot esculenta Crantz), their number and source in Tamil Nadu used for morphological diversity

S. No

ME181ME182 ME183ME185ME186ME187ME189ME193ME195ME199ME201ME204ME205ME207ME209ME211ME224ME232ME234ME238ME240ME242ME243ME245ME246ME247ME250ME251ME253ME254ME256ME259ME261ME263ME265ME267ME269ME278ME281ME282ME287ME290ME293ME295ME296ME297ME299ME306ME309ME310ME312ME313ME314ME315

Entry

SM 1444-2 (CIAT.9)SM 1444-3 (CIAT.10)SM 1444-4 (CIAT.11)SM 1444-6 (CIAT.13)SM 1444-7 (CIAT.14)SM 1444-8 (CIAT.15)SM 1450-2 (CIAT.17)SM 1456-3 (CIAT.22)SM 1456-5 (CIAT.23)SM 1670-4 (CIAT.30)SM 1670 (CIAT.32)SM 1765-3 (CIAT.36)SM 1765-4 (CIAT.370)SM 1765-6 (CIAT.39)SM 1774-1 (CIAT.41)SM 1774-3 (CIAT.44)SM 1783-4 (CIAT.57)SM 1784-1 (CIAT.69)SM 1785-5(CIAT.80)SM 1785-5 (CIAT.80)SM 1787-1 (CIAT.83)SM 1787-3 (CIAT.87)SM 1787-4 (CIAT.89)SM 1781-5 (CIAT.92)SM 1787-7 (CIAT.94)SM 1788-1 (CIAT.96)SM 1789-1, MCOL 1505 (CIAT.99)SM 1789-2 (CIAT.100)SM 1789-4 (CIAT.103)SM 1789-5 (CIAT.104)SM 1789-7 (CIAT.108)SM 1790,(CIAT.113)SM 1790-4(CIAT.116)SM 1791-2(CIAT.120)SM 1883-1(CIAT.122)SM 1883-3(CIAT.125)SM 1885-1(CIAT.128)SM 189 1-6,(CIAT.139)SM 1893-1(CIAT.142)SM 1893-2(CIAT.143)SM 8208-2(CIAT.151)CM.8208-5 (CIAT. 154)CM.8208-2 (CIAT. 151)SM.1670-2 (CIAT. 1590)CM.8487-1 (CIAT. 160)CM.8487-2 (CIAT. 161)SM. 1430-1(CIAT.164)SM. 1432-6(CIAT.171)SM. 1432-9(CIAT.174)SM. 1732-10(CIAT.175)SM. 1447-1(CIAT.1790)SM. 1447-2(CIAT.1810)SM. 1447-3(CIAT.182)SM. 1449-1(CIAT.1850)

Source

555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108

S. No

ME318ME319ME320ME321ME322ME323ME324ME325ME326ME328ME329ME330ME332ME333ME335ME336ME337ME343ME346ME347ME348ME349ME352ME353ME355ME356ME358ME360ME366ME368ME371ME373ME374ME375ME376ME377ME378ME380ME384ME385ME386ME395ME397ME398ME399ME401ME403ME406ME410ME411ME412ME421ME422ME423

Entry

SM. 1450-1(CIAT.190)SM. 1450-2(CIAT.192)SM. 1454-1 (CIAT.193)SM. 1454-2 (CIAT.196)SM. 1454-3 (CIAT.197)SM. 1454-4 (CIAT.198)SM. 1454-5 (CIAT.199)SM. 1454-6 (CIAT.200)SM. 1454-7 (CIAT.201)SM. 1454-9 (CIAT.203)SM. 1454-10 (CIAT.205)SM. 1454-11 (CIAT.206)SM. 1456-1 (CIAT.207A)SM. 1456-2 (CIAT.208)SM. 1456-41 (CIAT.210)SM. 1456-5 (CIAT.211)SM. 1456-6 (CIAT.212)SM. 1460-1 (CIAT.219)SM. 1460-4 (CIAT.224)SM. 1460-5 (CIAT.225)SM. 1460-6 (CIAT.226)SM. 1460-7 (CIAT.227)SM. 1461-3 (CIAT.231)SM. 1461-4 (CIAT.232)SM. 1470-1 (CIAT.234)SM. 1470-2 (CIAT.235)SM. 1472-2 (CIAT.237)SM. 1488-1 (CIAT.240)SM. 1490-4 (CIAT.248)SM. 1490-4 (CIAT.251)SM. 1491-4 (CIAT.254)SM. 1491-6 (CIAT.258)SM. 1519-1(CIAT.259)SM. 1519-1(CIAT.260)SM. 1520-1(CIAT.262)SM. 1520-2(CIAT.263)SM. 1520-3(CIAT.264)SM. 1520-5(CIAT.268)SM. 1521-4(CIAT.274)SM. 1521-5(CIAT.279)SM. 1521-6(CIAT.281)SM. 1521-15(CIAT.291)SM. 1523-2(CIAT.293)SM. 1523-3(CIAT.294)SM. 1532-1(CIAT.295)SM. 1532-1(CIAT.295)SM. 1532-1(CIAT.299)SM. 1532-8(CIAT.308)SM. 1542-4(CIAT.308)SM. 1542-5(CIAT.310)SM. 1542-6(CIAT.311)SM. 1542-15(CIAT.327)SM. 1543-1(CIAT.328)SM. 1543-2(CIAT.329)

Source

109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162

S. No

ME426ME427ME429ME430ME434ME438ME440ME444ME448ME450ME453ME454ME456ME458ME460ME464ME465ME474ME475ME476ME489ME491ME501ME514ME519ME528ME539ME541ME544ME550ME557ME559ME560ME566ME568ME569ME572ME574ME575ME576ME577ME578ME579ME580ME585ME590ME595ME598ME600ME604ME605ME606ME609ME610

Entry

SM. 1544-2(CIAT.332)SM. 1544-3(CIAT.332)SM. 1544-4(CIAT.335)SM. 1545-1(CIAT.336)SM. 1554-1(CIAT.340)SM. 1554-7(CIAT.348)SM. 1557-2(CIAT.350)SM. 1564-4(CIAT.354)SM. 1564-8 (CIAT.358)SM. 1564-10 (CIAT.360)SM. 1565-1 (CIAT.363)(CIAT.363 A)SM. 1565-3 (CIAT.365)SM. 1565-5 (CIAT.367)SM. 1573-2 (CIAT.369)SM. 1573-6 (CIAT.373)SM. 1573-7 (CIAT.374)SM. 1656-4 (CIAT.385)SM. 1665-1 (CIAT.386)SM. 1665-2 (CIAT.387)SM. 1667-4 (CIAT.400A)SM. 1667-6 (CIAT.402)SM. 1679-3 (CIAT.413)SM. 1754-2 (CIAT.429)SM. 1774-1 (CIAT.435)SM. 1778-2 (CIAT.444)SM. 17842 (CIAT.458)SM. 1784-4 (CIAT.461)SM. 1784-7 (CIAT.464)SM. 1784-13 (CIAT.470)SM. 1789-5 (CIAT.477)SM. 1789-7 (CIAT.479)SM. 1789-8 (CIAT.480)SM. 1789-14 (CIAT.486)SM. 1791-2 (CIAT.489)SM. 1791-2 (CIAT.490)SM. 1885-2 (CIAT.492 A)SM. 1885-4 (CIAT.494)SM. 1885-5(CIAT.495)SM. 1885-6(CIAT.496A)SM. 1858-1(CIAT.497)SM. 1858-2(CIAT.499)SM. 1859-1(CIAT.501)SM. 1860-1(CIAT.503)SM. 1864-1(CIAT.508)SM. 1865-3(CIAT.513)SM. 1948-1(CIAT.518)SM. 1948-4(CIAT.521)SM. 1948-6(CIAT.524)SM. 1953-1(CIAT.528)SM. 1953-2(CIAT.529)SM. 1960-1(CIAT.530)SM. 1969-1(CIAT.534)SM. 1969-2(CIAT.535)

Source

Cassava Genetic Diversity188

Suppl. Table 1. continued

163164165166167168172173174175176177178179180181182183184185186187188189190191192193194195

S. No

ME611ME613ME614ME615ME616ME624ME633ME635ME637ME638ME641ME643ME645ME647ME650ME654ME656ME657ME659ME660ME668ME681ME682ME687ME688ME690ME691ME692ME693ME695

Entry

SM. 1969-3(CIAT.536)OMR 36-05-1(CIAT 541)OMR 36-05-2(CIAT 542)OMR 36-05-3(CIAT 543)OMR 36-05-4(CIAT 544)OMR 36-05-12(CIAT 552)OMR 36-42-3(CIAT 563)OMR 36-42-1(CIAT 565A)OMR 36-42-3(CIAT 567)OMR 36-42-4(CIAT 569)OMR 36-42-7(CIAT 572)OMR 36-60-1(CIAT 575)OMR 36-67-1(CIAT 577)OMR 36-73-2(CIAT 582)OMR 36-74-1(CIAT 585)SM 1532 (CIAT 590)SM 1564 (CIAT 592)SM 1564 (RDS) (CIAT 594)SM 1504 (RDS) (CIAT 595)SM 1461 (RDS) (CIAT 596)SM 1460 (CIAT 608)Thondamuthur.1Thondamuthur.2Thondamuthur.7Thondamuthur.8Thondamuthur.10Thondamuthur.11Thondamuthur.12TCS HybridThondamuthur Hybrid.1-3

Source

196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225

S. No

ME696ME697ME698ME699ME700ME701ME702ME703ME704ME705ME706ME707ME708ME709ME710ME711ME712ME713ME714ME715ME716ME717ME718ME719ME720ME721ME722ME723ME724ME725

Entry

TCS HybridThondamuthur Hybrid.1-5TCS HybridThondamuthur Hybrid.1-5TCS Hybrid 4-8Thondamuthur Hybrid.1-5TCS Hybrid 4-10Thondamuthur Hybrid.1-5TCS Hybrid 4-12Thondamuthur Hybrid.1-5TCS Hybrid 4-14Thondamuthur Hybrid.1-5TCS Hybrid 5-1Thondamuthur Hybrid.1-5TCS Hybrid 5-5TCS 5-6TCS 5-7TCS 5-8TCS 5-9TCS 5-10TCS 5-11TCS 6-1TCS 6-2TCS 6-4TCS 6-6TCS 7-2TCS 7-4TCS 7-8TCS 10-1TCS 10-2

Source

226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255

S. No

ME727ME728ME729ME730ME731ME732ME733ME734ME735ME736ME737ME738ME739ME740ME741ME742ME743ME744ME745ME746ME747ME748ME749ME750ME751ME752ME753ME754ME755ME756

Entry

TCS 10-4TCS 10-5TCS 11-1TCS 11-3TCS 11-4TCS 11-5TCS 11-6TCS 11-7TCS 11-9TCS 11-11TCS 11-12TCS 24-1TCS 24-2TCS 24-3TCS 24-4TCS 24-8TCS 24-9TCS 24-10TCS 24-11TCS 24-12TCS 24-13TCS 25-1TCS 28-2TCS 28-5TCS 28-6TCS 28-8TCS 28-10TCS 28-11TCS 28-12TCS 28-13

Source

JCSB 2011 (September) 14 (3) : 179 ~ 189 189

Unaffected shoots, no symptoms

Mild chlorosis, mild distortions at bases of most leaves, while theremaining parts of the leaves and leaflets appear green and healthy

Pronounced mosaic pattern on most leaves, narrowing and distortion ofthe lower one-third of the leaflets

Severe mosaic distortion of two thirds of most leaves and generalreduction of leaf size and stunting of shoots

Very severe mosaic symptoms on all leaves, distortion, twisting, mis-shapen, and severe leaf reductions of most leaves accompanied bysevere stunting of plants

1

2

3

4

5

Suppl. Table 2. The damage scores used to determine the level of resist-ance to CMD

Symptom description Score

SSRY2SSRY28SSRY324

(CT)18CC(CT)16

(CT)26(AT)3AC(AT)2

(GC)3A(CA)5A(CA)2

CGCCTACCACTGCCATAAAC

TTGACATGAGTGATATTTTCTTGAG

CGCTTACAACACCACCTTCA

TGATGAAATTCAAAGCACCA

GCTGCGTGCAAAACTAAAAT

GCTTGATCTCAGCCATGTCA

167180206

0.460.660.63

Suppl. Table 3. SSR Locus, types of repeat, primers sequences, ampliconsize, and polymorphic information content (PIC)

Types of repeat

SSR Locus

Forward primer (5' - 3')

Reverse primer (5' - 3')

PIC Amplicon size (bp)