INTRODUCTION Guava (Psidium guajava) belongs to the family Myrtaceae and has a diploid chromosom number 2n = 22. Guava is believed to have originated in Central America (Hayes, 1953) and is now well-adapted to India, being widely grown throughout the country. Although guava is self pollinated, 35-40% cross pollination takes place in some cultivars, providing a heterogeneous, open- pollinated seedling population with adequate genetic variation (Pathak and Ojha, 1993). Most guava varieties have evolved through selection from seedling variants. Guava fruit is a rich source of vitamins (A and C), dietary fiber, carotenoids, essential oils and pectin. Besides its nutritional properties, the leaves and bark of P. guajava have a long history of medicinal use that still continues today (Joseph and Mini Priya, 2011). The present study aims to use microsatellite markers to measure genetic diversity in guava germplasm. This would help breeders choose genetically diverse germplasm for use as parents in breeding programmes to develop superior hybrids. 3Department of Biotechnology, Gokaraju Rangaraju Institute of Engineering and Technology, Hyderabad-500090, India. Assessment of genetic diversity in guava (Psidium guajava) germplasm using microsatellites M.V. Naga Chaithanya1, M.R. Dinesh2, C. Vasugi2, D.C. Lakshmana Reddy1, D. Sailaja3 and C. Aswath1* 1Division of Biotechnology, 2Division of Fruit crops ICAR- Indian Institute of Horticultural Research, Hesaraghatta Lake Post, Bengaluru-560 089, India *E-mail: aswath@iihr.ernet.in ABSTRACT Although the varietal diversity is fairly rich in guava, most varieties lack one or more desirable characters. Hence, attempts were made for improving specific traits, viz., attractive pink pulp colour, soft seeds, medium fruit size, high TSS and high ascorbic acid. Genetic diversity analysis is a prerequisite for identifying potential parents in breeding programs and germplasm conservation. Molecular characterization helps discriminate closely-related genotypes, as, this technique is unaffected by environment, rendering it more reliable. In this study, 48 polymorphic SSRs screened from a total of 115 SSR markers were used for analyzing marker segregation in 72 guava accessions. Statistical analysis was done using IDENTITY1.0 and CERVUS 3.0 software. Cluster analysis was done with DARwin 5.0 software, using Wards Minimum Variance method, and weighted group neighbour joining method, to check reliability of grouping among clusters. The trend in grouping was found to be similar in both methods. Dendrograms generated showed that the hybrids clustered with their parents; exotic collections fell into two different sub-groups based on productivity; the wild species formed one group; and Navalar cultivars from Dharwad clustered together, reflecting similar origin. Key words: Guava, genetic diversity, dendrogram, simple sequence repeats Microsatellites consist of tandemly repeated units of DNA, show high levels of allelic diversity per locus, are co-dominant in nature and highly reliable. Thus, these can be used for infering genotyping information, and are best suited for genetic diversity studies in both plants and animals (Cholastova and Knotova 2012). MATERIAL AND METHODS a. Plant material and molecular markers Guava germplasm maintained in the Field Gene Bank (FGB) at Indian Institute of Horticultural Research, Bengaluru, India, was selected for the study. The germplasm consisted of hybrids, exotic collections, local collections and wild species used as rootstocks or as parents in the disease resistance breeding program (Vasugi and Dinesh, 2007). Total genomic DNA of the 72 guava accessions was extracted from leaf material by Doyle and Doyle method (1990) with minor modifications. The precipitated DNA was dissolved in TE buffer and integrity of genomic DNA isolated was determined by electrophoresis in 0.8% agarose gel. DNA quantification J. Hortl. Sci. Vol. 9(2):117-125, 2014 118 was done using a Gene Quant UV- Spectrophotometer (GE Health Care Bio-sciences Ltd; England), and diluted accordingly. Out of a total number of 72 accessions (Table 1), 24 guava accessions (diverse with respect to fruit characteristics, viz., fruit weight, TSS-Total Soluble Solids, ascorbic acid, pulp colour, seed hardness and fruit weight) were initially selected. Genotyping was performed using the following polymerase chain reaction profile: Volume of the reaction mixture was 20µl which contained 1X buffer (10 mM Tris HCl of pH 8, 50mM KCl, 1.5mM MgCl2), 0.5µM of each primer, 200µM of dNTPs, 0.5 units of Taq DNA polymerase (GeNei, Bangalore) and 50ng genomic DNA. DNA amplification was done as per the PCR program of Risterucci et al (2005). Amplified PCR products were separated on 3% agarose gel (3B Black Bio Biotech India, Ltd.) loaded with 100bp ladder (Fermentas). The gels were stained with ethidium bromide and were photographed using UVIPRO platinum gel documentation unit. Molecular weight analysis of the amplified alleles was made in comparison with a 100 bp ladder loaded along with the samples by using UVITEC platinum ID software (ver.12, Cambridge, UK). Genotyping was done to screen 115 SSR (simple sequence repeats) markers, of which only 48 were found to be polymorphic. PCR amplification of the remaining accessions was performed using the 48 shortlisted polymorphic markers (Table 2) to obtain allelic molecular weight data of all the 72 accessions. b. Statistical and genetic diversity analysis Statistical parameters such as number of alleles per locus (k), allele frequencies, expected heterozygosity (EXP.HET.), observed heterozygosity (OBS.HET.) and Polymorphic Information Content (PIC) were analyzed using allele frequency analysis module of Cervus 3.0 (Kilinowski et al, 2007). Probability of identity (PI) was calculated using IDENTITY1.0 software. DARwin 5.0 software was applied to construct a dendrogram by both Wards Minimum Variance method and Neighbor Joining method. Factorial analysis was performed using DARwin 5.0 (http://darwin.cirad.fr) (Perrier et al, 2003). RESULTS AND DISCUSSION 1. Allelic diversity The 48 SSRs exhibited unique amplification fragments in the 72 accessions. Results of statistical analysis are depicted in Table 2. A total of 249 alleles were amplified. Amplicon size varied from 71bp (mPgCIR207) to 386bp (mPgCIR100). Based on results on statistical analysis (Table 2), the number of alleles ranged from 2 (mPgCIR029, mPgCIR236) to 11 (mPgCIR201), with a mean number of 5.33 alleles per locus (which was higher than 4.5 alleles per locus reported earlier) (Rodriguez et al, 2007). PIC values ranged from 0.260 (mPgCIR236) to 0.811 (mPgCIR243), with a mean of 0.563. Expected heterozygosity ranged from 0.078 (mPgCIR038) to 0.838 (mPgCIR243), with a mean of 0.616. PI values computed using IDENTITY1.0 software (Wagner and Sefc, 1999) ranged from 0.052 to 0.847 for the loci mPgCIR321 and mPgCIR038, respectively (which was higher than the range 0.031 to 0.487 reported earlier) (Kanupriya et al, 2011). 2. Genetic diversity analysis Cluster analysis was performed using the distance- based clustering method, which takes pair-wise distance matrix as an input for analysis, by a specific clustering algorithm (Johnson and Wichern, 1992). 1. Nasik 2. Local 1 3. Chittidar 4. Sindh 5. Local 2 6. Local 3 7. Behat Coconut 8. G6 9. EC-147037 10. CIW5 11. Nagpur Seedless 12. Abu Ishaqwala 13. Phili Pink 14. Lucknow- 42 15. CIW1 16. Apple Colour 17. Aneuploid-2 18. Dhareedar 19. CIW2 20. GR1 21. Surkha Chitti 22. Aneuploid- 1 23. Chakaiya Ruthamnagar 24. Surkha Chitti Neptuani 25. Arka Amulya 26. Arka Mridula 27. Allahabad Safeda 28. Safed Jam 29. C.P.A. White 30. Ben Dror 31. Smooth Green 32. Benaras 33. Portugal 34. Lalit 35. Mirzapur Seedling 36. Sardar Guava 37. Florida Seedling 38. Seedless Triploid 39. Karela 40. EC-147039 41. Spear Acid 42. Purple Local 43. Red Flesh 44. Pati 45. Dharwad 46. Hisar Safeda 47. Superior Sour Lucidum 48. White Flesh 49. Psidium molle 50. Psidium cattelianum 51. Psidium quadrangularis 52. Psidium chinensis 53. Psidium friedrichsthalianum 54. S.P.No.6 55. S.P.No.7 56. 7-12 EC147036 57. Hafsi 58. Bangalore Local 59. 7-39147034 60. 9-35147036 61. EC-162904 62. Parker Dessert 63. Sabdana Badri 64. Kohir Long 65. Kohir Safeda 66. Swetha 67. CISH-G-1 68. Local White 69. Beaumont 70. Kg Guava 71. Thailand Guava 72. Kamsari Table1. List of accessions used in the present study Naga Chaithanya et al J. Hortl. Sci. Vol. 9(2):117-125, 2014 119 Table 2: List of polymorphic markers with results of statistical analysis Sl. Locus F=Forward primer (5’-3’) Expected size No. of Allele OBS. EXP. PIC PI No. R=Reverse primer (3’-5’) (bp) alleles size (bp) HET HET 1 mPgCIR339 F: CCGAAGACGAGGAGATTA 160 5 140-227 0.070 0.658 0.582 0.181 R: TTAAGTGGAAAATCACAGTTG 2 mPgCIR243 F: ACAGCAGGACACAAAGGA 174 7 107-212 0.292 0.798 0.764 0.072 R: GCTCTGAGGTGGTTTTCAT 3 mPgCIR182 F: GAGGAAGAAACCCGAAGTTA 181 8 87-202 0.264 0.8 0.769 0.068 R:GGTAGAAAGATCGGAAAGAC 4 mPgCIR236 F: ACTCATATTCCGTTTGCATC 164 2 154-168 0.056 0.301 0.254 0.507 R:GAATTAACGACGAGTTCCAC 5 mPgCIR316 F: GCTTCATATTACAAACCTTGG 232 4 192-281 0.239 0.464 0.416 0.317 R:GATCTAACTGACTTGCCAAAA 6 mPgCIR326 F: AGAACAAGACACGAGAAGAG 116 6 83-179 0.250 0.787 0.748 0.081 R:AAAATCTACGCACAAACC 7 mPgCIR207 F: CAAGATTTGCCTCAAGAAAC 136 5 71-145 0.306 0.460 0.433 0.319 R:AACTAAATAGCCTGCTGGTG 8 mPgCIR206 F:GGAAGTTTCAAAGTAACAGCAC 181 6 174-295 0.111 0.764 0.721 0.096 R:AGAATGAGTCCATGCTCAAA 9 mPgCIR220 F:AGAGCAGTGGTTGCTATTTT 218 7 145-164 0.083 0.731 0.679 0.121 R:CCCATCTCTTACTTTTCTTGTG 10 mPgCIR277 F:AGCCGATTATGATTACCTG 173 5 144-191 0.250 0.732 0.689 0.113 R:CGATTCACTCCCTCATTACT 11 mPgCIR039 F:GCTCACCTTACTCATTCAGC 155 4 145-200 0.014 0.524 0.406 0.309 R:CTGTTGCTAAGAGCTTTCGT 12 mPgCIR222 F:CCAGAATCAGACATAGTTAGAG 166 3 169-213 0.171 0.393 0.337 0.381 R:CTGAAGACATCAACATGGAA 13 mPgCIR093 F:GCATCATGTGTTTGAACGAT 123 6 102-168 0.194 0.803 0.768 0.070 R:AAGTGTGCGTTCTCCATCT 14 mPgCIR099 F:TCAAAGTCCAAAACTCATGC 220 4 194-267 0.208 0.532 0.475 0.276 R:GGGATGGAGTAAAGATGAAA 15 mPgCIR042 F:CTCACCCAAAATCTACACAAG 107 3 110-140 0.029 0.322 0.296 0.436 R:AAGGGACTGGACGATGTT 16 mPgCIR100 F:CTAGAAGTCGAAGAATGGAA 128 5 122-386 0.239 0.671 0.617 0.154 R:TTTGTTAGTATCGGAGTCGAG 17 mPgCIR185 F:AACGCATCTGGCATTGAT 117 4 97-135 0.141 0.308 0.285 0.476 R:CCTTGGTCTCCCTCTTACTC 18 mPgCIR165 F:TAAGGGATTCATTTCCGAGT 124 3 127-176 0.029 0.523 0.411 0.289 R:CTGGTGTGACGATGACTTTT 19 mPgCIR029 F:CTCGCTTCAATCTCCATCTA 162 2 166-202 0.521 0.414 0.326 0.409 R:AGCGACACAGACTCTTCATT 20 mPgCIR154 F:CTTCAGCTACAGCCTTTCC 138 8 102-285 0.903 0.794 0.759 0.074 R:GGAGAAAGCAGAAATTCCA 21 mPgCIR038 F:AGCCTGTTTTACGCCTTC 111 3 102-131 0.028 0.081 0.079 0.847 R:CGGCTGCTCTATTGTTATTT 22 mPgCIR194 F:GCAGAGAATCGAAGCACTA 172 6 154-208 0.278 0.747 0.695 0.113 R:GCAAGCACAGGTTCTACTTT 23 mPgCIR193 F:GAACGTGGGTTACATACCAT 122 4 102-132 0.028 0.594 0.506 0.252 R:ATCACCGTCCTCCTAAATCT 24 mPgCIR027 F:AGCACTTAGGGACAAATTCA 292 4 262-337 0.167 0.668 0.598 0.178 R:CTCACTCTCCTCCATTCAAG 25 mPgCIR191 F:GACCCTCCCACTTATATTTTG 210 6 216-282 0.485 0.766 0.726 0.071 R:AAGCTGACATAACAGTCGAA 26 mPgCIR091 F:GCGGTGGATTTGAATTTAG 125 3 107-142 0.324 0.552 0.465 0.272 R:CCAAGTAACCCACAACAATA 27 mPgCIR031 F:TCTCACTGATGCAACTTTTC 128 8 104-191 0.159 0.616 0.580 0.156 R:CCCATTTTCATCTCAAAGTC 28 mPgCIR157 F:AACCACCAAACCATACACC 209 4 163-224 0.246 0.692 0.636 0.128 R:CGACCAACCCTACATTCTG Assessment of genetic diversity in guava using microsatellites J. Hortl. Sci. Vol. 9(2):117-125, 2014 120 Table 2: Contd. Sl. Locus F=Forward primer (5’-3’) Expected size No. of Allele OBS. EXP. PIC PI No. R=Reverse primer (3’-5’) (bp) alleles size (bp) HET HET 29 mPgCIR161 F:TCTCAAGGACCAACAAGAAG 246 5 218-283 0 0.681 0.622 0.136 R:AGGACTTAGCTTGGGTTTTC 30 mPgCIR111 F:CAACCTCGTTTGAGTCTTCT 115 5 86-156 0.222 0.420 0.394 0.363 R:AACATCATTGGGACCATTC 31 mPgCIR041 F:AAGTGGTGTCAGCAACTACC 136 4 130-170 0.014 0.579 0.505 0.214 R:CTTAGTTTGACCGCTCCAGT 32 mPgCIR174 F:GCCACTGTGTAAGAGGATTG 261 4 181-273 0.108 0.626 0.545 0.158 R:ATTGTGGGAGATTGGAGAC 33 mPgCIR184 F:AAGCTACAATCGACGAAAAC 221 5 171-254 0.171 0.706 0.660 0.117 R:CACTATTAGCGAACCTGCAT 34 mPgCIR104 F:ATTCCCGTGGATTATGTATC 120 2 125-141 0 0.423 0.332 0.381 R:ACAACCATTTTCTCCTCATC 35 mPgCIR109 F:AATTTCCACAGATCACAAGG 110 5 104-147 0.083 0.686 0.620 0.162 R:GGCATCTCCATCAAATACAT 36 mPgCIR325 F:AAACGCTCGAATCAGTTG 172 7 140-202 0.319 0.807 0.773 0.068 R:CCAAGAAACACAGGGATTAC 37 mPgCIR200 F:CCTTGCTTTGGTGAGGTC 178 8 141-377 0.203 0.685 0.645 0.118 R:GCTAATTCAGTCCTTCCAACT 38 mPgCIR321 F:TTTTGGCCTGGGAATATAG 129 8 114-191 0.209 0.810 0.778 0.052 R:TAAAACGAAAGCAGAAAACC 39 mPgCIR032 F:CGCCTTTCGTAAAAGAAGT 100 5 73-136 0.071 0.672 0.613 0.148 R:TCATATACTCGGACAAAACG 40 mPgCIR102 F:AATTGGTGTAGCATCTGGA 176 5 181-255 0.403 0.661 0.586 0.188 R:GCCTACCATGAACAGAGAAA 41 mPgCIR201 F:TTTGCCTTCGAGCTTCTACT 133 11 120-303 0.471 0.791 0.754 0.070 R:ACAATTTCGTGGGCTCGT 42 mPgCIR203 F:ATGAAGGCATTACCTAAGAC 126 3 127-341 0.086 0.547 0.447 0.274 R:ACCCTATTAACCCTTAGCAA 43 mPgCIR205 F:ACCTCTCCAGCTCTACACG 101 5 89-164 0.458 0.786 0.745 0.084 R:GAGGTTGTCGAAGGTTGAT 44 mPgCIR098 F:CATCAACTTTCCAGGCATA 127 4 116-148 0.0 0.663 0.595 0.154 R:CCATTCAGTCGGTTTGAC 45 mPgCIR101 F:ATGGCTGTAAGAAGCAAAAG 110 5 100-164 0.074 0.413 0.368 0.317 R:GAAGAAATGTAGGTGCGTTC 46 mPgCIR150 F:CCTAGTGACTCGAAGCAATC 108 5 106-152 0.153 0.657 0.592 0.182 R:TTGAGCCCTAGCATAGACAG 47 mPgCIR133 F:CGATCTTGGAATGTAAGAGG 148 8 134-244 0.181 0.709 0.659 0.133 R:TGGATTTGCAGGTTCTATCT 48 mPgCIR437 F:ACAACAGTTCTGATCCCAAA 153 6 155-346 0.099 0.725 0.678 0.113 R:CTCGGAGACACAGAGGTCTA bp= number of base pairs, OBS. HET= Observed heterozygosity, EXP. HET= Expected heterozygosity, PIC= polymorphic information content , PI= probability of identity a Wards Minimum Variance method Wards Minimum Variance method (Ward, 1963) generates a graphic representation such as a tree or a dendrogram, in which clusters can be visually identified. Confidence limits of different clades in the dendrogram were tested by bootstrapping 1000 times to assess repetitiveness of genotype clustering (Felsenstein, 1985) in both the methods. Dendrogram generated by Wards Minimum Variance method is shown in Figure 1. This method showed two clusters: a major cluster (Cluster 1) with 50 accessions, and a minor cluster (Cluster 2) with 22 accessions. Subgroup-C1 included genotypes like Sindh, Chittidar, Hafsi, Behat Coconut, Local 1, Nasik, Local 3, Bangalore Local, CIW5, Abu Ishakwala, and Nagpur Seedless. Sub group-C2 included seven varieties and five wild species, along with Phili Pink and Lucknow-42. Most of the exotic collections grouped with Local-2 in Group-D, leading to the inference that Local-2 is an introduction. Fourteen accessions were clustered in Subgroup-E1 as shown in Figure 1. Genotypes like Dhareedar, Aneuploid -2 and Apple Colour clustered with CIW2 and CIW1, in Subgroup-E2. Local - White, S.P. No.7, Bendror, CISH-G-1, Swetha, Beaumont, S.P. No.6 clustered in Group-F of Cluster-1. Naga Chaithanya et al J. Hortl. Sci. Vol. 9(2):117-125, 2014 121 Cluster-2 had one Group-G which was divided into Sub-groups G1 and G2. In Subgroup-G1, all the white-pulped varieties clustered with Purple Local, a purple-pulp accession. Many pink-pulp varieties clustered with two white-pulp accessions, Florida Seedling and Superior Sour Lucidum, in Subgroup-G2. b. Neighbor Joining method (NJ) To overcome systemic errors, an alternative method known as Neighbor Joining is used in phylogenetic studies. This removes the assumption that the data are ultrameric (Swofford et al, 1996). In this method, bootstrapping values of the allele frequencies can be displayed to assess reliability of the nodes. The dendrogram obtained is shown in Fig. 2. The Neighbor Joining method is discussed in detail, as, it includes bootstrapping. The bootstrap values varied from 2% to 100%. Highest bootstrap value of 100% was observed for the varieties Arka Amulya and Kohir Long. A similar low bootstrapping value, ranging from 3% to 100%, was reported earlier by Hadadinejad et al (2011). Cluster analysis clearly showed that the accessions fell broadly into three clusters. Cluster-1 was divided into Group-A and Group-B. Group-A was subsequently subdivided into A1 and A2. A1 Sub-group clustered 11 accession together showing that these were genetically closer; but morphologically, the similarity was not visible, perhaps due to minor differences in allelic composition. Clustering of five wild species with Phili Pink and Lucknow- 42 in Subgroup-A2 represents their close genetic similarity; it can be inferred indirectly that these are quite dissimilar to the commonly cultivated Psidium guajava. Similar clustering of wild species was earlier reported by Rajkumar et al, (2011). Group-B consisted of five exotic collections, with higher productivity (EC-162904, G6, 9-35147036, 7-39147034, EC-147037) (Vasugi and Rami Reddy, 2003). Cluster-2 with 26 germplasm accessions was divided into Group-C and Group-D. Group-C was further divided in to Sub-groups C1 and C2, that clustered 21 and one genotypes, respectively. Cluster C1 confirmed the parentage of hybrids (Dinesh and Vasugi, 2010) as Arka Mridula, and the hybrid Arka Amulya clustered with its maternal parent, Allahabad Safeda. Other hybrids (Kohir Safeda and Safed Jam) clustered with their parent, Allahabad Safeda. In C2, only GR1 was present. Navalur varieties grouped with Dhareedar, Aneuploid-2 and Apple Colour, indicating their genetic similarity within Group-D. Cluster-3 was divided into Sub- groups E1 and E2. Subgroup-E1 was further divided into Group-F and G. Group-F was sub-divided into F1 and F2. F1 was divided into F1a and F1b and contained white and pink pulp varieties in two different clusters, respectively. Subgroup-E2 clustered three accessions, Hisar Safeda, Purple Local and White Flesh, which separated from the other 19 genotypes. All the remaining 19 genotypes under E1 were divided into four Sub-groups like F1 (F1a, F1b, and F2) and G. Subgroup-F1 consist of both white and pink pulp varieties in two different clusters. Further clustering resulted in clear differentiation of the remaining pink and white pulp varieties. The white-pulp varieties clustered in Sub-group F1a. Sub-group F1b grouped all pink pulp varieties with Florida Seedling, a white pulp accession. Sub-groups F2 clustered two pink pulp varieties with Superior Sour Lucidum, a white-pulp accession. This grouping may perhaps be due to their highly acidic nature. In Sub-group G, pink-pulp varieties like Thailand Guava and Pati clustered together. Similar differentiation of white and pink pulp varieties was reported by Kanupriya et al (2011). c. Factorial analysis Factorial analysis represented in Fig. 3 is a type of Principal Co-ordinate Analysis used for deriving a 2-3 dimensional scatter plot of individuals. This method facilitates identification of individuals showing intermediacy between two groups (Lessa, 1990). Individuals belonging to a single plot reveal sets of genetically similar individuals (Karp et al, 1997). The picture consists of X axis and Y axis, based on which it is divided into four co-ordinates (Co-1, Co-2, Co-3, and Co-4). Interestingly, the accessions that grouped in Cluster 1 in Neighbor Joining method (NJ) were included in Co-2 (except P.quadrangularis, which grouped in Co-1). This alignment of P. quadrangularis in Co-1 of factorial analysis may be due to the superior morphological traits like high stamen number, large flower and fruit, and good flavour, compared to the other species used in the study (Vasugi and Dinesh, 2007). Accession G6 showed intermediacy between Co-1 and Co-2. Co-4 included accessions from Cluster-2 in NJ method, except Dhareedar, Aneuploid-2 and Apple Colour (which grouped in Co-2). Co-1 and Co-3 together included all the accessions belonging to cluster 3 in Neighbour Joining method. Irrespective of the method used, pattern of clustering among genotypes was found to be similar. Both the cluster-analysis methods grouped individuals into stringently defined groups or clusters. Finally, factorial analysis clearly confirmed the patterns obtained by cluster analysis. Assessment of genetic diversity in guava using microsatellites J. Hortl. Sci. Vol. 9(2):117-125, 2014 122 Fig. 1. Dendrogram generated using Wards Minimum Variance method from the computed genetic distances of simple matching coefficient. The black dot on the left of the dendrogram indicates the origin, and the line at the bottom indicates the coefficient of Jaccards Dissimilarity Matrix. Naga Chaithanya et al J. Hortl. Sci. Vol. 9(2):117-125, 2014 123 Fig. 2. Dendrogram generated using Weighted Neighbor Joining method from the computed genetic distances of simple matching coefficient. Bootstrap values supporting nodes are shown on the branches. The black dot on the left of the dendrogram indicates origin, and the line at the bottom indicates coefficient of Jaccards Dissimilarity Matrix. From the dendrograms, it can be deduced that many accessions were comparable to known superior varieties, and, can be used as parents in future guava breeding programs. Allelic pattern shown by the primer combinations evaluated in the present study confirms the high discriminatory capacity of SSR markers. Therefore fingerprinting of guava accessions can be done using such data. An acceptable level of genetic diversity was detected, as, a number of clusters were formed, allowing for efficient selection of parents in future breeding programs. Parentage was also confirmed through molecular diversity analysis. Exotic collections have a good demand in the processing Assessment of genetic diversity in guava using microsatellites J. Hortl. Sci. Vol. 9(2):117-125, 2014 124 Fig. 3. Factorial analysis generated using DARWin 5.0 software Naga Chaithanya et al J. Hortl. Sci. Vol. 9(2):117-125, 2014 125 industry because of their pink pulp, with sweet-acid blend and high productivity. These can be exploited in guava breeding programmes in India. ACKNOWLEDGEMENT The authors are indebted to Council for Scientific and Industrial Research (CSIR) for financial help as a Fellow to the first author, and Director, IIHR for providing facilities for this study. REFERENCES Cholastova, T. and Knotova, D. 2012. Using morphological and microsatellite (SSR) markers to assess the genetic diversity in alfalfa (Medicago sativa L.). Int’l. J. Biol. Vet. Agril. Food Engg., 6:146-152 Dinesh, M.R. and Vasugi, C. 2010. Guava improvement in India and future needs. J. Hortl. Sci., 5:94-108 Doyle, J.J. and Doyle, J.L. 1990. Isolation of plant DNA from fresh tissue. Focus, 12:13-15 Felsenstein, J. 1985. Confidence limits on phylogenies: an approach using the bootstrap. Evolution, 39:783-791 Hadadinejad, M., Ebadi, A., Naghavi, M.R. and Nikkhah, R. 2011. Genealogy and molecular diversity of Iranian grapevine progenies. J. Agril. Sci. Tech., 13:1147-1161 Hayes, W.B. 1953. Fruits growing in India. Kitabistan, Allahabad, India Johnson, A.R. and Wichern, D.W. 1992. Applied multivariate statistical analysis. 3rd edition, Prentice-Hall, Englewood Cliffs, NJ, USA Joseph, B. and Mini Priya, R. 2011. Review on nutritional, medicinal, pharmacological properties of guava (Psidium guajava L.). Int’l. J. Pharm. Biosci., 2:54-69 Kalinowski, S.T., Taper, M.L. and Marshall, T.C. 2007. Revising how the computer program Cervus accommodates genotyping error increases success in paternity assignment. Mol. Ecol., 16:1099-1106 Kanupriya, Madhavi Latha, P., Aswath C., Reddy, L., Padmakar, B., Vasugi, C. and Dinesh, M.R. 2011. Cultivar identification and genetic fingerprinting of guava (Psidium guajava L.) using microsatellite markers. Int’l. J. Fruit Sci., 11:184–196 Karp, A., Kresovich, S., Bhat, K.V., Ayad, W.G. and Hodgkin, T. 1997. Molecular tools in plant genetic resources conservation: A guide to the technologies IPGRI, Rome Lakshmana Reddy, D. C., Kiran, K., Srinivas Reddy, S.H., Kanupriya, Aswath, C. and Singh, T.H. 2012. SSR- based DNA barcodes as a tool for identification of eggplant genotypes. Int’l. J. Veg. Sci., 18:260-271 Lessa, E.P. 1990. Multidimensional analysis of geographic genetic structure. Sys. Biol., 39:242-252 Mahajan, R., Tabia, S., Raina, G. and Mangotra, N. 2012. Assessment of genetic diversity of non-basmati rice of Jammu and Kashmir using microsatellite markers. J. Cereals and Oil Seeds, 3:21-27 Mondini, L., Noorani, A. and Pagnotta, M.A. 2009. Assessing plant genetic diversity by molecular tools. Diversity, 1:19-35 Pathak, R.K. and Ojha, C.M. 1993. Genetic resources of guava. Adv. Hort. Fruit Crops., 1:143-147 Perrier, X., Flori, A. and Bonnot, F. 2003.Genetic diversity of cultivated tropical plants (ed. P. Hamon). Data Analysis Methods, 43–76 Rajkumar, G., Weerasena, J., Fernando, K. and Liyanage, A. 2011. Assessment of genetic diversity among Sri Lankan rice varieties by AFLP markers. Pl. Genetic Resources, 9:224-228 Risterucci, A.M., Duval, M.F., Rohde, W. and Billotte, N. 2005. Isolation and characterization of microsatellite loci from Psidium guajava L. Mol. Ecol. Notes, 5:745-748 Rodriguez, N.N., Valdés-Infante, J., Becker, D., Velázquez, B., Coto, O., Ritter, E. and Rohde. 2004. Morphological, agronomic and molecular characterization of Cuban accessions of guava (Psidium guajava L.). J. Genet. Breed., 8:70-90 Swofford, D.L., Olsen, G.J., Waddell, P.J. and Hillis, D.M. 1996. Phylogenetic inference. In: D.M. Hillis et al. (eds.) Molecular Systematics, 2nd edition, Sinauer Associates, Sunderland, MA., USA, pp. 407–514 Vasugi, C. and Rami Reddy, P.V. 2003. Chacterization and evaluation of some exotic guava accessions under Bangalore conditions. Indian J. Pl. Genet. Resources, 17:189-192 Vasugi, C. and Dinesh, M.R. 2007. Genetic variability in some Psidium species. Indian J. Agril. Sci., 77:420- 423 Wagner, H.W. and Sefc, K.M. 1999. Identity 1.0, Centre for Applied Genetics, University of Agricultural Sciences, Vienna, Austria Ward, J.H. Jr.1963. Hierarchical grouping to optimize an objective function. J. Amer. Statistical Assoc., 58:236–244 (MS Received 06 October 2014, Revised 29 November 2014, Accepted 03 December 2014) Assessment of genetic diversity in guava using microsatellites J. Hortl. Sci. Vol. 9(2):117-125, 2014