RELIABILITY OF THE RAPD TECHNIQUE FOR GERMPLASM

ANALYSIS OF SESAME (Sesamum indicum L) FROM VENEZUELA

Bertha Salazar, Hernán Laurentín, Martha Dávila and Miguel A. Castillo

Bertha Salazar. Agronomical Engineer, Universidad Centroccidental Lisandro Alvarado (UCLA), Venezuela. Technician, Laboratory of Molecular Embriology, UCLA, Venezuela. e-mail: berthasalco@yahoo.com

Hernán Laurentín. Agronomical Engineer, UCLA, Venezuela. M.Sc. in Agronomy, Universidad Central de Venezuela (UCV). Professor, UCLA, Venezuela. e-mail: helauren@yahoo.com

Martha Dávila. Biologist, UCV, Venezuela. M.Sc. in Horticulture, UCLA, Venezuela. Ph.D. in Agronomy, University of Nebraska-Lincoln, USA. Professor, UCLA, Venezuela. Address: Departamento de Ciencias Biológicas, Universidad Centroccidental Lisandro Alvarado, Lara, Venezuela. e-mail: martad@ucla.edu.ve

Miguel A. Castillo. Biologist, Universidad Simón Bolívar, Venezuela. M.Sc. and Ph.D. in Biology, Instituto Venezolano de Investigaciones Científicas. Professor, UCLA, Venezuela. e-mail: macastillo@ucla.edu.ve

SUMMARY

The random amplified polymorphic DNA (RAPD) technique was used to evaluate two commercial cultivars (Fonucla and UCLA-1) and 7 lines (UCLA-249; UCLA-295; UCLA-37-1; UCLA-65; UCLA-83; UCLA-90; UCV-2) of sesame (Sesamum indicum L) obtained by the breeding program of the Universidad Centroccidental Lisandro Alvarado, Venezuela. Ninety four polymorphic bands using 12 random 10-mer primers indicated a high level of variability. Three primers were able to discriminate 8 out of the 9 materials and, in combination, they were able to resolve all of them. Probability of identical match by chance was 5.22×10^-29. Polymorphic information content (PIC), resolving power (RP) and marker index (MI) of each primer failed to correlate significantly with number of genotypes resolved. Unique bands were observed in 6 genotypes with 9 primers. Jaccard’s similarity coefficients ranged from 0.04 to 0.53. UPGMA clustering resulted in two major groups and the genotype classification agreed closely with the grouping observed when using principal coordinates analysis (PCA). The results of the present study showed that RAPD-based fingerprinting was a useful tool to identify unequivocally the sesame genotypes and to assess the genetic variability of the breeding stocks.

CONFIABILIDAD DE LA TÉCNICA RAPD PARA EL ANÁLISIS DE GERMOPLASMA DE AJONJOLÍ (Sesamum indicum L.) DE VENEZUELA

RESUMEN

La técnica de ADN polimórfico amplificado al azar (RAPD) fue utilizada en dos cultivares comerciales (Fonucla y UCLA-1) y 7 líneas (UCLA-249; UCLA-295; UCLA-37-1; UCLA-65; UCLA-83; UCLA-90; UCV-2) de ajonjolí (Sesamum indicum L) obtenidas por el programa de mejoramiento genético de la Universidad Centroccidental Lisandro Alvarado, Venezuela. Noventa y cuatro bandas polimórficas derivadas del uso de 12 deca-iniciadores indicaron un alto nivel de variabilidad. Tres de los iniciadores discriminaron 8 de los 9 materiales y, en combinación, pudieron resolverlos a todos. La probabilidad de que ocurrieran coincidencias idénticas fue de 5,22×10^-29. El contenido de información de polimorfismo (CIP), el poder de resolución (PR) y el índice del marcador (IM) de cada iniciador no se correlacionaron de manera significativa con el número de genotipos resueltos. El coeficiente de similitud de Jaccard varió de 0,04 a 0,53. El agrupamiento realizado mediante UPGMA resultó en dos grupos principales relacionándose estrechamente con los grupos observados mediante el análisis de coordenadas principales (CP). Estos resultados demuestran que la técnica de RAPD es una herramienta útil para la identificación inequívoca de genotipos de ajonjolí y para evaluar la variabilidad de materiales genéticos usados en programas de mejoramiento genético.

CONFIABILIDADE DA TÉCNICA RAPD PARA A ANÁLISE DE GERMOPLASMA DE GERGELIM (Sesamum indicum L.) DA VENEZUELA

RESUMO

A técnica de ADN polimórfico amplificado aleatoriamente (RAPD) foi utilizada em dois cultivares comerciais (Fonucla e UCLA-1) e 7 linhas (UCLA-249; UCLA-295; UCLA-37-1; UCLA-65; UCLA-83; UCLA-90; UCV-2) de gergelim (Sesamum indicum L) obtidas pelo programa de melhoramento genético da Universidade Centroccidental Lisandro Alvarado, Venezuela. Noventa e quatro faixas polimórficas derivadas do uso de 12 deca-iniciadores indicaram um alto nível de variabilidade. Três dos iniciadores discriminaram 8 dos 9 materiais e, em combinação, puderam resolvê-los a todos. A probabilidade de que ocorresem coincidencias idênticas foi de 5,22×10^-29. O conteúdo de informação de polimorfismo (CIP), o poder de resolução (PR) e o índice do marcador (IM) de cada iniciador não se correlacionaram de maneira significativa com o número de genotipos resolvidos. O coeficiente de similitude de Jaccard variou de 0,04 a 0,53. O agrupamento realizado mediante UPGMA resultou em dois grupos principais relacionando-se estreitamente com os grupos observados mediante a análise de coordenadas principais (CP). Estes resultados demonstram que a técnica de RAPD é uma ferramenta útil para a identificação inequívoca de genotipos de gergelim e para avaliar a variabilidade de materiais genéticos usados em programas de melhoramento genético.

KEYWORDS / DNA Markers / Genetic Variability / RAPD / Sesamum indicum /

Received: 09/30/2005. Modified: 03/27/2006. Accepted: 04/12/2006.

Introduction

Sesame is one of the most ancient oil seed crops (Bedigan and Harlan, 1986; Ashri, 1998). It is cultivated on 7 million ha world wide (FAO, 2005), mainly in developing countries of Africa, Asia and America (Montilla and Terán, 1996). Sesame seeds contain 50-60% oil of great stability due to the presence of natural antioxidants such as sesamin and sesamol (Brar and Ahuja, 1979; Ashri, 1989). The seed is a source of protein and is used for confections and for decorating bread and cakes (Uzun et al., 2003). In Venezuela, sesame is a rotation crop growing after maize (Zea mays L) during the dry season in the western llanos, mainly in Turén, Portuguesa State (Laurentín et al., 2003). Harvested areas in the past 3 years have remained stable at nearly 60000ha, with an average production of 30000Mg per year (FAO, 2005), which is mostly exported to major customers such as USA and England (Laurentín et al., 2003).

Venezuelan sesame production has been strongly supported by national breeding programs. Germplasm characterization is present throughout all the phases in plant breeding. Before a breeding program is established, it is necessary to evaluate genetic diversity in the species, and afterwards it is desirable to describe unequivocally the new cultivars. Genetic characterization in crop species, including sesame, can be determined using morphological and agronomic traits, isozyme, and DNA marker analysis (Isshiki and Umezaki, 1997; Karp et al., 1997; Parani et al., 1997; Díaz et al., 1999; Rao, 2004).

Within DNA markers, the random amplified polymorphic DNA (RAPD) technique is suitable when a medium level of polymorphism is expected. Additionally, this technique has advantages such as low development cost, low level of training required, and low cost per assay (Karp et al., 1997), especially important for countries of lesser resources. RAPD does not require previous knowledge of genome sequences because arbitrary primers are used. RAPD markers have been used for diversity analysis in a vast array of crops, including sesame (Bhat et al., 1999; Nanthakumar et al., 2000; Ercan et al., 2004) and for the identification of cultivars (Fernández et al., 2002; Archak et al., 2003; Rajora and Rahman, 2003). The purpose of the present study was to evaluate and compare the informative capacity of RAPD markers in 9 sesame lines, and to determine the genetic relationships among them.

Materials and methods

Plant material

Six samples (UCLA-249, UCLA-295, UCLA-37-1, UCLA-65, UCLA-83 and UCLA-90) from the sesame breeding program at the Universidad Centrooccidental Lisandro Alvarado, two commercial varieties known as UCLA-1 (Montilla and Terán, 1996) and ‘Fonucla’ (Montilla and Cedeño, 1991) and one individual selection from the Arawaca cultivar termed UCV-2 (data not published), were used for this study and were grown in the field during 2002. All the genotypes are "white seeded", except UCV-2 which is dark seeded.

DNA isolation

Petals of flowers from several plants were collected and stored at -84ºC. Total genomic DNA was extracted using the protocol recommended by GenomicPrep Cells and Tissue DNA Isolation Kit^® of Amershan Pharmacia Biotech (USA). The DNA concentration was estimated with a calibration curve using DNA concentrations standards (l Ladder DNA, Operon Technologies Inc., USA ) in 2% agarose gel at 90V for 30min in TAE buffer (0.04M Tris-acetate, 0.001M EDTA, pH 8.0) and staining with ethidium bromide.

DNA amplification

Genomic DNA was amplified using the protocol Ready to Go RAPD Analysis Beads^® from Amersham Pharmacia Biotech (USA). Each bead contained AmpliTaq^® DNA polymerase, dNTPs (0.4mM of each for 25µl reaction volume), BSA (2.5µg) and PCR buffer (3mM MgCl₂, 30mM KCl and 10mM Tris, pH 8.3, in 25µl reaction volume). PCR reaction was carried out in a DNA Thermal Cycler^® 2400 (Perkin Elmer, USA.). Each 25µl reaction mixture contained one bead, 5µl of template DNA, 5µl of primer (Operon Technologies, Alameda, USA.) and 15µl Milli-Q ultrapure water (resistivity <1MW·cm^-1). The PCR amplification conditions were: initial extended step of denaturation at 95ºC for 1min followed by 45 1min cycles of denaturation at 94ºC for, primer annealing at 36ºC for 1min and elongation at 72ºC for 2min, followed by an extended elongation step at 72ºC for 4min. Reaction products were mixed with 2µl of 10X loading dye (0.25% bromophenol blue, 0.25% xylene cyanol and 40% sucrose, w/v) and spun briefly in a microfuge before loading. The amplification products and 100bp DNA ladder (Operon Technologies Inc., USA) were electrophoresed on 2% agarose gel at 100V, followed by staining with ethidium bromide and photographed on Polaroid 667 film under UV light. A negative control was utilized with all RAPD reactions. The list of used primers used and their sequences is given in Table I.

Data analysis

All amplifications were repeated twice and only reproducible bands were considered for analysis. The amplified fragments were named by primer and their size in base pairs (bp). Each band or amplification product was considered as a RAPD marker. The bands were recorded as present (1) or absent (0) in order to construct a binary matrix of RAPD phenotypes.

The discriminatory power of RAPD markers was evaluated by three parameters. The polymorphic information content (PIC) for each RAPD marker was calculated, as proposed by Roldán-Ruiz et al. (2000), as PIC_i = 2f_i(1-f_i), where PIC_i is the polymorphic information content of marker i, fi is the frequency of the marker bands present, and (1-fi) is the frequency of absent marker bands. Dominant markers as RAPD have a maximum of 0.5 when half of the accessions have the band and the other half does not have the band (De Riek et al., 2001). PIC was averaged over the bands for each primer.

The resolving power (Rp) of the twelve primers was calculated according to Prevost and Wilkinson (1999) as Rp = SIb where Ib (band informativeness) takes the value of 1-[2x|0.5-p|], p being the proportion of the genotypes containing the band. The third parameter used was the marker index (MI) as proposed by Powell et al. (1996) and used by Milbourne et al. (1997). MI is the product between diversity index (equivalent to PIC) and effective multiplex ratio (EMR), where EMR is defined as the product of the fraction of polymorphic loci and the number of polymorphic loci. This parameter was calculated for each primer.

To study the informativeness of RAPD in sesame, the number of different fingerprints (NF) per primer and the number of genotypes with unique fingerprint (NGUF) per primer were recorded. Pearson correlation was performed between the number of different fingerprints per primer and the three parameters evaluated, as well as between the number of lines with unique fingerprint per primer and the three parameters.

To obtain the level of confidence in identifying the 9 materials, the probability of identical match by chance (Pi) was calculated, as proposed by Wetton et al. (1987) and Ramakrishna et al. (1994), as Pi = Xⁿ, where X is a similarity index between 2 genotypes, expressing the probability that a band present in one of them is also present in the other, and n is the average number of bands in the two genotypes being compared. X was calculated as the ratio 2N_AB/(N_A+N_B), where N_AB is the number of bands present in the genotypes, N_A total number of bands in genotype A and N_B is the total number of bands in genotype B. This probability was calculated for each primer, and also over all the primers by multiplying the individual probabilities.

Jaccard´s similarity coefficient values for each pair-wise comparison between genotypes were calculated and a similarity coefficient matrix was constructed. This matrix was subjected to the unweighted pair-group method for arithmetic averages analysis (UPGMA) in order to generate a dendrogram. Cophenetic correlation was calculated as a measure of the faithfulness of cluster analysis (Rohlf and Sokal, 1981). In addition, the data was subjected to principal coordinates analysis (PCA; Sneath and Sokal, 1973). All the numerical taxonomic analyses were conducted using the software NTSYS-PC, version 2.11T (Exeter Software, New York). The robustness of the nodes in the dendrogram was tested by bootstrap analysis using the Winboot program developed by IRRI.

Results

Table I summarizes the amplification products resulting for each primer and the values of the calculated parameters explained in the previous section.

The total number of bands obtained was 94, being all of them polymorphic. The number of bands per genotype ranged between 12 for UCLA37-1 and 53 for UCLA-249, with an average of 36 bands per genotype. The number of amplification products per primer ranged from 5 (OPC-02, OPC-13, OPM-13) to 13 (OPC-07). Twenty-two unique bands were obtained in 6 genotypes with 9 primers, where the materials UCLA-65 and UCLA-295 presented the highest number of unique bands with 6 and 8, respectively. The primers OPA-01, OPE-08 and OPM-06 identified both the highest number of fingerprints (8) and the highest number of genotypes with unique fingerprint (7), but they could not distinguish all the 9 lines. For instance, primer OPA-01 did not resolve UCLA-65 and Fonucla, OPE-08 did not resolve UCLA-83 and UCLA-37-1, and primer OPM-06 did not resolve UCLA-90 and UCV-2. The PIC, with average of 0.37, ranged from 0.17 (primer OPC-02) to 0.45 (primer OPC-19), having primers OPC-19, OPM-06, and OPE-08 the highest PIC values. The highest Rp values were obtained with the primers OPE-07, OPA-01 and OPM-06, and highest MI values with the primers OPE-07, OPA-01, and OPC-07. No significant correlation was found neither between number of fingerprints nor lines with exclusive fingerprints, with PIC, Rp or MI.

The probability of identical match by chance (Pi) ranged between 0.0002 (OPE-07) and 0.0314 (OPM-13), and the total probability obtained by multiplying all the probabilities was 5.22×10^-29.

Jaccard’s similarity coefficients ranged from 0.036 to 0.53, with an average of 0.2835 (Table II). The UPGMA-based phenogram consisted of two clusters; the first one grouped UCLA-37-1 and UCLA-83, and the second one included the rest (Figure 1). The cophenetic correlation coefficient was 0.86. Bootstrapping analysis resulted in at least 89% of confidence limits for the two major clusters. The PCA showed that the first three coordinate axes accounted for 40.88% of the total variation (Figure 2). The pattern agreed closely with the grouping observed in the UPGMA-based phenogram.

 

Discussion

Accurate identification of germplasm carried out by means of DNA fingerprints is a useful tool for checking the identity and purity of a variety. In the present study, sesame lines could be identified based on RAPD fingerprints using twelve 10-mer primers, with an extremely low probability of getting an identical match by chance. Primers OPA-01, OPE-08 and OPM-06 are considered of high value for fingerprint in sesame because each one was able to resolve 8 of the 9 materials studied. These three primers should be useful for fingerprinting, because when combined they are able to resolve all the 9 lines. Pi for only these three primers was 2.73×10^-9, which is a very low probability of identical match by chance.

RAPD fingerprint showed a total of 22 unique bands, which have a high potential value since they can be converted into STS (sequence tagged site) markers. This is useful for detecting mixes between cultivars (Fernández et al., 2002). Similarly, Bhat et al. (1999) suggested that, in sesame, the primer OPM-06 could be useful for obtaining cultivar/genotype specific profiles. Therefore, exclusive bands could be converted to sequence-characterized amplified regions (SCARs). The same suggestion could be applied for the nine primers in the present study that resulted in unique bands.

The high level of polymorphism reported in this study is comparable to 87% of polymorphism reported by Bhat et al. (1999). They also reported 100% of polymorphism for primers OPA-02, OPA-19 and OPM-06. Ercan et al. (2004) reported a high level of polymorphism (78%) studying Turkish sesame accessions.

Studies about the discriminatory power of RAPD primers have not been carried out in sesame, but similar studies have been done in other self pollinated crops such as barley. Russell et al. (1997) obtained a PIC of 0.52, higher than the 0.37 value reported in this study and Fernández et al. (2002) reported an Rp value of 3.85, similar to 4.26 in this study. However, the values by themselves do not have a clear meaning. More studies have been carried out with cross pollinated crops, but few of them compare parameters such as PIC, Rp, and MI with number of fingerprints or number of resolved genotypes. When the identification of cultivars is the purpose of the fingerprints, the most important function of a primer is to discriminate as many cultivars as possible, this aspect being the most important to consider. Prevost and Wilkinson (1999) and Fernández et al. (2002) found a strong and linear relationship between the ability of a primer to distinguish genotypes and Rp, but not with MI. The data from Rajora and Rahman (2003) indicated significant correlation (P<0.05) between PIC and number of genotypes observed, but not with number of cultivars with unique genotype. The lack of correlation observed in this study and the lack of consistency in other studies suggest that it would be better to consider how many genotypes are discriminated by a primer, instead of calculating parameters such as PIC, Rp and MI.

The range of Jaccard´s similarity coefficients among lines was rather ample, with the maximum value (0.53) very low when compared with the results obtained by Bhat et al. (1999).

The UPGMA-based phenogram and PCA showed a similar pattern, in spite of the low percentage of variation accounted by the first three axes of PCA. Similar relationships between both analyses have been reported by Ercan et al. (2004). In this study, both analysis displayed lines UCLA-37-1 and UCLA-83 as the most divergent. This finding does not agree with the study by Laurentín et al. (2004), where yield components were analyzed on the same lines used in the present study (except UCV-2) by means of PCA. The only common feature in both graphics was the grouping of UCLA-90, UCLA-249 and UCLA-1. The discrepancy can be explained by the differences in the methodology applied, since molecular characterization covers only the genome variability (Ovesná et al., 2002) excluding the environmental influence (Rao, 2004), while morphological characterization, mainly of quantitative traits as studied by Laurentín et al. (2004), were subjected to strong environmental influence (Karp et al., 1997; Rao, 2004). However, it is interesting to note that 33% of the line were also grouped together in both studies. Therefore, they should not be the most suitable genotypes to obtain diversity when recombined. UCLA-37-1 or UCLA-83 appeared as the most suitable and possible parents for an eventual new "white seed" population, if crossed with some of the other materials studied. The information obtained from both of the above analyses could not be correlated with the origin of these genotypes, since this is unknown.

The results of the present study demonstrate that RAPD-based fingerprints is a useful tool to identify unequivocally sesame genotypes. This information can be used successfully for cultivar identification and for assessing the genetic variability of breeding stocks.

Acknowledgements

The authors thank the CDCHT (Consejo de Desarrollo Científico, Humanístico y Tecnológico of Universidad Centroccidental Lisandro Alvarado for the economic support.

REFERENCES

1. Archak S, Gaikwad B, Gautam D, Rao E, Swamy K and Karihaloo J (2003) DNA fingerprinting of Indian cashew (Anacardium occidentale L.) varieties using RAPD and ISSR techniques. Euphytica 230: 397-404.        [ Links ]

2. Ashri A (1989) Sesame. In Roebblen G, Downey RK, Ashri A (Eds) Oil crops of the World. McGraw Hill. New York, NY, USA. pp.375-387.        [ Links ]

3. Ashri (1998) Sesame breeding. Plant Breed Rev 16: 179-228.        [ Links ]

4. Bedigan D, Harlan JR (1986) Evidence for the cultivation of sesame in the ancient world. Econ Bot 40: 137-154.        [ Links ]

5. Bhat VK, Babrekar P, Lakhanpaul S (1999) Study of genetic diversity in Indian and exotic sesame (Sesamum indicum L.) germplasm using random amplified polymorphic DNA (RAPD) markers. Euphytica 110: 21-33.        [ Links ]

6. Brar GS, Ahuja KL (1979) Sesame: its culture, genetics, breeding and biochemestry. In Malik CP (Ed.). Annu. Rev. Plant Sci.. Kalyani. New Dehli, India. pp. 245-313.        [ Links ]

7. De Riek J, Calsyn E, Everaert I, Van Bockstaele E, De Loose M (2001) AFLP based alternatives for the assessment of distinctness, uniformity and stability of sugar beet varieties. Theor Appl Genet 103: 1254-1265.        [ Links ]

8. Díaz AJ, Layrisse A, Pugh T (1999) Análisis de la diversidad genética en ajonjolí mediante electroforesis de isoenzimas. Agron Tropical (Maracay) 49: 169-186.        [ Links ]

9. Ercan GA, Taskin M, Turgut K (2004) Analysis of genetic diversity in Turkish sesame (Sesame indicum L.) populations using RAPD markers. Genet. Resour. Crop Evol. 51: 599-607.        [ Links ]

10. FAO (2005) FAOstat Databases. http://apps.fao.org/cgi-bin/nph-db.pl        [ Links ]

11. Fernández M, Figueiras A, Benito C (2002) The use of ISSR and RAPD markers for detecting DNA polymorphism, genotype identification and genetic diversity among barley cultivars with known origin. Theor. Appl. Genet. 104: 845-851.        [ Links ]

12. Isshiki S, Umezake T (1997) Genetic variations of isozymes in cultivated sesame. Euphytica 93: 375-377.        [ Links ]

13. Karp A, Kresovich S, Bhat K, Ayad W, Hodgkin T (1997) Molecular tools in plant genetic resources conservation: a guide to the technologie. IPGRI Tech. Bull. Nº 2. International Plant Genetic Resources Institute. Rome, Italy. 47 pp.        [ Links ]

14. Laurentín H, Pereira C, Sanabria M (2003) Phytochemical Characterization of six Sesame (Sesamum indicum L.) genotypes and their relationships with resistance against the sweetpotato whitefly Bemisia Tabaci Gennadius. Agron. J. 95: 1577-1582.

15. Laurentín H, Montilla D, García V (2004) Relación entre el rendimiento de ocho genotipos de ajonjolí (Sesamum indicum L.) y sus componentes. Comparación de metodologías. Bioagro 16: 153-162.

16. Milbourne D, Meyer R, Bradshaw J, Baird E, Bonar N, Provan J, Powell W, Waught R (1997) Comparison of PCR-based marker systems for the analysis of genetic relationships in cultivated potato. Mol. Breed. 3: 127-136.

17. Montilla D, Cedeño T (1991) Fonucla. Nueva variedad de ajonjojlí. Bioagro 3: 52-54.

18. Montilla D, Terán H (1996) UCLA-1, una nueva variedad de ajonjolí (Sesamum indicum L.) Bioagro 8: 26-29.

19. Nanthakummar G, Singh KN, Vaidyanatan P (2000) Relationships between cultivated sesame (Sesamun sp.) and the wild relatives based on morphological characters, isozymes and RAPD markers. J. Genet. Breed. 54: 5-12.

20. Ovesná J, Poláková K, Leisová L (2002) DNA analyses and their applications in plant breeding. Czech J. Genet. Plant Breed. 38: 29-40.

21. Powell W, Margenta M, Andre C, Hanfrey M, Vogel J, Tingey S, Rafalsky A (1996) The utility of RFLP, RAPD, AFLP and SSR (microsatellite) markers for germplasm analysis. Mol. Breed. 2: 225-238.

22. Prevost A, Wilkinson M (1999) A new system of comparing PCR primers applied to ISSR fingerprinting of potato cultivars. Theor. Appl. Genet. 98: 107-112.

23. Rajora O, Rahman M (2003) Microsatellite DNA and RAPD fingerprinting, identification and genetic relationships of hybrid poplar (Populus x canadiensis) cultivars. Theor. Appl. Genet. 106: 470-477.

24. Ramakrishna W, Lagu M, Gupta V, Ranjekar P (1994) DNA fingerprinting in rice using oligonucleotide probes specific for simple repetitive DNA sequences. Theor. Appl. Genet. 88: 402-406.

25. Rao N (2004) Plant genetic resources: advancing conservation and use through biotechnology. African J. Biotechnol. 3: 136-145.

26. Rohlf F, Sokal R (1981) Comparing numerical taxonomic studies. System. Zool 30: 459-490.

27. Roldán-Ruiz I, Dendauw J, VanBockstaele E, Depicker A, De Loose M (2000) AFLP markers reveal high polymorphic rates in ryegrasses (Lollium spp.) Mol. Breed. 6: 125-134.

28. Russell J, Fuller J, Macaulay M, Hatz B, Jahoor A, Powell W, Waugh R (1997) Direct comparison of levels of genetic variation among barley accessions detected by RFLPs, AFLPs, SSRs and RAPDs. Theor. Appl. Genet. 95: 714-722.

29. Sneath PHA, Sokal RR (1973) Numerical Taxonomy. Freeman. San Francisco, CA, USA. 517 pp.

30. Uzun B, Lee D, Donini P, Cagirgan MI (2003) Identification of a molecular marker linked to the closed capsule mutant trait in sesame using AFLP. Plant Breed. 122: 95-97.

31. Wetton J, Carter R,  Parkin D, Walters D (1987) Demographic study of a wild house sparrow population by DNA fingerprinting. Nature 327: 147-149.