Abstract
• Premise of the study: To investigate the genetic diversity of a root crop, Xanthosoma sagittifolium, and to facilitate germplasm conservation, microsatellite loci were developed and characterized by genotyping 39 accessions from different geographic origins.
• Methods and Results: Using a microsatellite-enriched library approach, 17 polymorphic microsatellite markers were identified and characterized. The number of alleles for each locus ranged from two to six. Observed and expected heterozygosities ranged from 0.00 to 0.97 and from 0.09 to 0.78, respectively. Additionally, cross-amplification of these microsatellite markers was tested successfully in other species of Xanthosoma and Caladium, with rates varying from 23.5% to 100%.
• Conclusions: These results indicate the effectiveness of microsatellite loci developed for the characterization of X. sagittifolium genetic diversity. They are crucial for the future investigation of population dynamics and clonal identification and, therefore, for prioritizing germplasm conservation. They should also enable research on other related species.
Keywords: Caladium lindenii, genetic conservation, microsatellite markers, polymorphism, tuber crop, Xanthosoma sagittifolium
Xanthosoma sagittifolium (L.) Schott (Araceae) is a monocotyledon aroid native to the tropical Americas, but its original place of domestication is still unknown. It is widely distributed throughout tropical regions in Africa, Southeast Asia, and Oceania. It is an allogamous species cultivated exclusively by vegetative propagation, preventing any possible genetic recombination. Its adaptive capacity is therefore almost nonexistent, and it is vulnerable to various pathogens such as dasheen mosaic virus (DMV) or Pythium myriotylum Drechsler (Lebot, 2009). Genetic resources are therefore of great value and need to be carefully identified, conserved, and protected. To date, the molecular genetic diversity of this plant has been investigated using only RAPD markers (Schnell et al., 1999; Offei et al., 2004) within a limited number of accessions and across restricted areas. The management of ex situ collections and the accurate identification of clones are often hampered by the lack of efficient markers. Present knowledge on wild crop relatives, population genetics, and spatial distribution of this species is insufficient and limited. It is therefore necessary to develop highly polymorphic codominant markers in X. sagittifolium.
Microsatellites or simple sequence repeats (SSRs) are highly polymorphic markers that are codominant and widespread in the genome. Such characteristics make them useful for a large range of applications in genetics. Consequently, they have been developed for a large number of plant species.
Here, we present the first set of polymorphic nuclear microsatellite markers suitable for germplasm diversity studies and further genetic conservation in X. sagittifolium. Additionally, the microsatellite primers that gave good results were tested for cross-amplification in related Xanthosoma and Caladium species.
METHODS AND RESULTS
Total genomic DNA was extracted from a silica gel–dried leaf sample following the protocol of Risterucci et al. (2000) and purified on NucleoBond PC20 columns (Macherey-Nagel, Düren, Germany). Genomic DNA of one accession of X. sagittifolium (living accession Xs10 at the Vanuatu Agricultural Research and Technical Centre [VARTC], Vanuatu) was restricted with RsaI (Invitrogen, Carlsbad, California, USA) and then used to construct a (GA)n and (GT)n microsatellite-enriched library following the protocol of Billotte et al. (1999). The enriched microsatellite fragments were then cloned into pGEM-T Easy Vector (Promega Corporation, Madison, Wisconsin, USA) as indicated by the supplier and were used to transform Escherichia coli DH10B competent cells (Invitrogen). Overall, 288 white transformed clones then underwent PCR amplification in a 50-μL reaction mixture containing a bacterial colony, 0.16 mM dNTP, 0.2 μM for each M13 primer, 50 mM KCl, 10 mM Tris-HCl, 0.001% glycerol, 2 mM MgCl2, and 1 unit Taq DNA polymerase. The PCR program was: initial denaturation at 95°C for 4 min; 30 cycles at 94°C for 30 s, 52°C for 45 s, and 72°C for 1 min 30 s; and final elongation at 72°C for 8 min. PCR products were then electrophoresed on 1.2% agarose gels at 80 V.
Out of 85% of clones giving positive amplification, 120 inserts were purified on NucleoSpin Gel and PCR Clean-up columns (Macherey Nagel), then sequenced using M13 primer and Big Dye Terminator version 3.1 (Applied Biosystems, Foster City, California, USA), and run on ABI 3500XL Genetic Analyzers (Applied Biosystems). The vector and adapter flanking sequences were eliminated, SSR presence was detected, and PCR primers were designed from a subset of 30 appropriate sequences using the SAT Web application (Dereeper et al., 2007). The primers were commercially synthesized (Sigma-Aldrich, Dorset, United Kingdom) with forward primers having an M13-tail added to their 5′ end (5′-CACGACGTTGTAAAACGAC-3′). The added M13-tails were labeled with IRD700 or IRD800 fluorochromes. To test the efficiency of the markers, a sample of 39 genomic DNAs of X. sagittifolium from three different geographic origins (20 from Vanuatu, 18 from Burkina Faso, and one from India) was used. The accessions are kept as living material at the Vanuatu Agricultural Research and Technical Center (Vanuatu accessions), at the University of Ouagadougou (Burkina Faso accessions), and at Central Tuber Crops Research Institute (India accessions) (Appendix 1). PCR amplification was performed in a Techne TC-412 Thermal Cycler as described by Chaïr et al. (2010). Allele number and sizes were determined using AFLP Quantar Pro 1.0 software (KeyGene, Wageningen, The Netherlands). Expected and observed heterozygosities were estimated using GENETIX 4.04 software (Belkhir et al., 2002), and polymorphism information content (PIC) values were estimated using CERVUS 3.0.3 software (Kalinowski et al., 2007).
Seventeen out of the 30 loci were identified as polymorphic and generated consistent amplification products. The characteristics of these new markers are summarized in Table 1. The number of alleles observed for each locus ranged from two (mXsCIR05, mXsCIR07, and mXsCIR14) to six (mXsCIR11 and mXsCIR20), with an average of 3.65 alleles per locus. The observed (Ho) and expected (He) heterozygosities ranged from 0.00 (mXsCIR05) to 0.97 (mXsCIR11 and mXsCIR22) (average: 0.61) and 0.09 (mXsCIR05) to 0.78 (mXsCIR11) (average: 0.53), respectively. Twelve markers presented Ho values higher than the He values, imparting an excess of heterozygous individuals in the studied sample and suggesting high rates of asexual reproduction, mainly due to vegetative propagation. In the absence of random mating, the tests for Hardy–Weinberg equilibrium and linkage disequilibrium were not performed. The PIC values obtained ranged from 0.01 (mXsCIR05) to 0.07 (mXsCIR11) (average: 0.48), with values greater than 0.5 for nine loci.
Table 1.
Characteristics of 17 polymorphic microsatellite markers isolated from Xanthosoma sagittifolium.
| Locus | Repeat motif | Clone size (bp) | Primer sequence (5′–3′) | GenBank accession no. | Ta (°C) | A | Allele size range (bp) | He | Ho | PIC |
| mXsCIR05 | (CA)8(CACA)3 | 196 | F: GCGCATTATTAACGAATATC | KF693219 | 50 | 2 | 212–216 | 0.0973 | 0.0000 | 0.099 |
| R: GTCATCTATGGCTATCACCT | ||||||||||
| mXsCIR07 | (TG)7(AG)19 | 204 | F: GGACTGGGAGTCTGAGTAG | KF693220 | 50 | 2 | 224–241 | 0.2778 | 0.2308 | 0.225 |
| R: CCTTTCCCCTCACTATAAA | ||||||||||
| mXsCIR10 | (AG)22 | 222 | F: GATGTCTGTAGTGGCCTAGT | KF693221 | 50 | 5 | 208–251 | 0.6586 | 0.5526 | 0.584 |
| R: AATTAAGTTGGGTGGTAGAT | ||||||||||
| mXsCIR11 | (TG)10(GA)16 | 204 | F: AATTCTTAGCAGCATTGTTA | KF693222 | 50 | 6 | 204–236 | 0.7883 | 0.9744 | 0.747 |
| R: CATTCGTATCAACTTCCTTT | ||||||||||
| mXsCIR12 | (TC)17(TTC)7 | 242 | F: TACATTTCCATTGCCATC | KF693223 | 50 | 4 | 223–282 | 0.4316 | 0.4359 | 0.389 |
| (TCCC)3(TTCTTG)3 | R: CAAATTAAAGAGGGAGACAG | |||||||||
| mXsCIR13 | (CA)8(AG)16 | 209 | F: GTTTCCTTTATTCGTTGATG | KF693224 | 50 | 5 | 223–244 | 0.7699 | 0.7949 | 0.739 |
| R: GTAGTGGCTGAGAATTGAAA | ||||||||||
| mXsCIR14 | (AG)20 | 196 | F: TACCCTACATTTGGGATCT | KF693225 | 50 | 2 | 190–216 | 0.1841 | 0.2051 | 0.178 |
| R: TTTTGGCTTTAGGTCTATTC | ||||||||||
| mXsCIR16 | (AG)15 | 183 | F: CTTATTGATGCCGAGAATAC | KF693226 | 50 | 3 | 202–205 | 0.6091 | 0.2308 | 0.531 |
| R: TTCCTCACAATATGTTCTCAT | ||||||||||
| mXsCIR19 | (AC)8(AC)24(AC)8 | 229 | F: CAACTTGTGTATCCTACATCC | KF693227 | 50 | 4 | 184–204 | 0.6769 | 0.8947 | 0.619 |
| R: GCGTGGTTTATGTGTATCTT | ||||||||||
| mXsCIR20 | (CT)11(TC)15(TCTA)3 | 220 | F: CCCTTATTGCTGTTTTCA | KF693228 | 50 | 6 | 226–247 | 0.6483 | 0.9487 | 0.586 |
| R: CATATCTCTTCCTCTCACCA | ||||||||||
| mXsCIR21 | (AG)30 | 194 | F: CTTAACCTTGTCAGCCTCT | KF693229 | 50 | 4 | 194–226 | 0.6295 | 0.8421 | 0.577 |
| R: GAGCGGTATAACAATTCATC | ||||||||||
| mXsCIR22 | (AG)22 | 211 | F: CGTGAGAAACACCTGAATTA | KF693230 | 50 | 4 | 216–226 | 0.6972 | 0.9744 | 0.629 |
| R: AATTTGCTCTGTCATTGTG | ||||||||||
| mXsCIR23 | (GA)23 | 166 | F: TGTAGGTATGGACACATGG | KF693231 | 50 | 3 | 176–184 | 0.5513 | 0.7436 | 0.475 |
| R: TTAAGACAAACCCTCAGC | ||||||||||
| mXsCIR24 | (AG)23 | 204 | F: AATTTGAAGTGAAACGATCA | KF693232 | 50 | 3 | 206–225 | 0.5362 | 0.4324 | 0.474 |
| R: TTCCTGTCATCAGAATTGTA | ||||||||||
| mXsCIR26 | (TC)9(TC)9 | 266 | F: TTCACCATTACTTGTCCACT | KF693233 | 50 | 3 | 245–285 | 0.5914 | 0.8974 | 0.511 |
| R: TTAACATGGGAACGTATCTT | ||||||||||
| mXsCIR27 | (AG)15(GAA)6 | 198 | F: TGCATGAATTGAAGAAAT | KF693234 | 50 | 3 | 217–245 | 0.4467 | 0.5385 | 0.385 |
| R: AACAAAGAGTCTCACCACAT | ||||||||||
| mXsCIR28 | (GA)9 | 201 | F: ACAGAAGTTGACATGGAGAG | KF709527 | 50 | 3 | 218–223 | 0.5017 | 0.7105 | 0.427 |
| R: AATGTTAAAGAGCAAAAGGA |
Note: A = number of observed alleles per locus; F = forward sequence; He = expected heterozygosity; Ho = observed heterozygosity; PIC = polymorphism information content; R = reverse sequence; Ta = optimized annealing temperature.
Additionally, cross-amplifications were carried out to test marker transferability to Caladium lindenii (André) Madison, previously classified as Xanthosoma, and 16 Xanthosoma species (14 referenced and two unreferenced) namely: X. atrovirens K. Koch & C. D. Bouché, X. blandum Schott, X. brasiliense (Desf.) Engl., X. ceronii Croat & L. P. Hannon, X. granvillei Croat & S. A. Thomps., X. harlingii Croat & L. P. Hannon, X. hylaeae Engl. & K. Krause, X. mexicanum Liebm., X. piquambiense sp. nov. Croat, Scherber. & G. Ferry, X. poeppigii Schott, X. pubescens Poepp., X. robustum Schott, X. violaceum Schott, X. viviparum Madison, X. sp. Croat, and X. sp. nov. Croat (Appendix 2), using the experimental protocol described above (Table 2). One out of 17 primers (mXsCIR11) showed 100% cross-amplification. The remaining markers were restricted to certain species, with rates varying from 23.5% (mXsCIR27) to 94.1% (mXsCIR12), with an average of 58.1%. Interestingly, a high number of markers (rates varying from 88.2% to 94.1%) showed transferability to the species X. brasiliense, X. robustum, X. violaceum, X. atrovirens, and X. blandum, suggesting high phylogenetic proximity of these species with X. sagittifolium. The remaining species seemed to be moderately close to X. sagittifolium as suggested by a lower transferability of the markers, with a rate varying from 35.3% to 58.8%. DNA of C. lindenii was amplified by only two markers (mXsCIR11 and mXsCIR23), confirming its classification in the Caladium genus as demonstrated by Loh et al. (2000).
Table 2.
Cross-amplification of 17 SSR loci from Xanthosoma sagittifolium in other species of Xanthosoma and Caladium.a
| Locus | C. lindenii (0-u-4195) | X. violaceum (2004) | X. atrovirens (013188) | X. blandum (080119) | X. brasiliense (070064) | X. ceronii (070377) | X. granvillei (031506) | X. harlingii (020201) | X. hylaeae (011463) | X. mexicanum (060617) | X. piquambiense sp. nov. (120412) | X. poeppigii (070769) | X. pubescens (031520) | X. robustum (060616) | X. viviparum (031522) | Transferability of SSR marker loci (%) |
| mXsCIR05 | — | + | + | + | + | + | — | + | — | + | — | + | — | + | + | 66.7 |
| mXsCIR07 | — | + | + | + | + | + | + | — | — | — | + | + | — | + | — | 60.0 |
| mXsCIR10 | — | + | + | + | + | — | — | + | — | — | — | — | — | + | — | 40.0 |
| mXsCIR11 | + | + | + | + | + | + | + | + | + | + | + | + | + | + | + | 100 |
| mXsCIR12 | — | + | + | + | + | + | + | + | + | + | + | + | + | + | + | 93.3 |
| mXsCIR13 | — | + | + | + | + | — | — | — | — | — | — | — | — | + | — | 33.3 |
| mXsCIR14 | — | + | + | + | + | + | + | + | + | — | + | + | — | + | + | 80.0 |
| mXsCIR16 | — | + | + | + | + | — | — | — | — | — | — | — | — | + | — | 33.3 |
| mXsCIR19 | — | + | + | + | + | + | + | + | + | — | + | + | + | + | + | 86.7 |
| mXsCIR20 | — | + | + | + | + | — | — | — | — | + | + | — | — | + | — | 46.7 |
| mXsCIR21 | — | + | + | + | + | — | — | — | — | — | — | — | — | + | — | 33.3 |
| mXsCIR22 | — | + | + | + | + | — | — | — | — | — | — | — | — | + | — | 33.3 |
| mXsCIR23 | + | + | + | + | + | + | + | + | + | + | — | + | + | + | — | 86.7 |
| mXsCIR24 | — | — | — | — | — | + | + | — | — | — | — | — | — | + | + | 26.7 |
| mXsCIR26 | — | + | + | + | + | + | + | — | — | + | — | + | + | + | + | 73.3 |
| mXsCIR27 | — | — | — | — | + | — | + | + | — | — | — | — | — | — | — | 20.0 |
| mXsCIR28 | — | + | + | + | + | + | + | + | + | + | — | + | + | + | + | 86.7 |
Note: + = successful amplification; — = failed amplification.
Voucher numbers are given in parentheses.
CONCLUSIONS
We present a first attempt to develop SSR molecular markers for X. sagittifolium. The 17 polymorphic markers identified here represent powerful tools for investigating genetic diversity, population genetic structure, and conservation biology. In addition, they are useful for clone and provenance identification—a necessary prerequisite for germplasm maintenance and the development of core collections. This research will considerably improve our knowledge of wild and cultivated cocoyam relationships. It will also serve as a base for the development of conservation strategies for this neglected crop. In addition, this set of markers proved to have broad transferability with other related species of the same genus. The information gained in the current study is therefore essential for future research on these species.
Appendix
Appendix 1.
Xanthosoma sagittifolium specimens used in this study. All specimens were conserved as living accessions at the Vanuatu Agricultural Research and Technical Centre (VARTC), Vanuatu; at the University of Ouagadougou, Burkina Faso; and at the Central Tuber Crops Research Institute (CTCRI), India.
| Collection no. | Locality (GPS coordinates) |
| MZN 28 | Batié, Burkina Faso (09°59.496′N, 002°55.978′W) |
| BDG 11a | Bérégadougou, Burkina Faso (10°43.393′N, 004°44.780′W) |
| BDG 11b | Bérégadougou, Burkina Faso (10°43.393′N, 004°44.780′W) |
| MDD 1 | Middebdo, Burkina Faso (09°59.258′N, 003°08.482′W) |
| TNS 1 | Tansié, Burkina Faso (10°51.717′N, 003°14.135′W) |
| CAD 1a | Cascade Bo, Burkina Faso (10°43.213′N, 004°44.745′W) |
| CAD 1b | Cascade Bo, Burkina Faso (10°43.213′N, 004°44.745′W) |
| CAD 1 | Cascade Bo, Burkina Faso (10°43.213′N, 004°44.745′W) |
| CAD 2 | Cascade, Burkina Faso (10°43.261′N, 004°44.650′W) |
| CAD 3a | Séréfédougou, Burkina Faso (10°43.401′N, 004°42.089′W) |
| CAD 3b | Séréfédougou, Burkina Faso (10°43.401′N, 004°42.089′W) |
| DTA 1 | Diatara, Burkina Faso (09°52.813′N, 002°55.980′W) |
| CAD 4a | Cascade Bo, Burkina Faso (10°43.305′N, 004°44.801′W) |
| CAD 4b | Cascade Bo, Burkina Faso (10°43.305′N, 004°44.801′W) |
| GUR 1 | Guiéri, Burkina Faso (10°58.384′N, 005°02.149′W) |
| CAD 5a | Cascade Bo, Burkina Faso (10°44.717′N, 004°43.226′W) |
| CAD 5b | Cascade Bo, Burkina Faso (10°44.717′N, 004°43.226′W) |
| CAD 6 | Banfora, Burkina Faso (10°36′N, 004°45′W) |
| Xs1 | Vanuatu (unknown) |
| Xs2 | Vanuatu (unknown) |
| Xs3 | Vanuatu (unknown) |
| Xs4 | Vanuatu (unknown) |
| Xs5 | Vanuatu (unknown) |
| Xs6 | Vanuatu (unknown) |
| Xs7 | Vanuatu (unknown) |
| Xs8 | Vanuatu (unknown) |
| Xs9 | Vanuatu (unknown) |
| Xs10 | Vanuatu (unknown) |
| Xs Hyb 1 | Vanuatu (unknown) |
| Xs Hyb 102 | Vanuatu (unknown) |
| Xs Hyb 157 | Vanuatu (unknown) |
| Xs Hyb 180 | Vanuatu (unknown) |
| Xs Hyb 184 | Vanuatu (unknown) |
| Xs Hyb 210 | Vanuatu (unknown) |
| Xs Hyb 227 | Vanuatu (unknown) |
| CTCRI_IND_19 | India (unknown) |
Appendix 2.
Xanthosoma and Caladium voucher specimens used in this study. All specimens were deposited at the botanical gardens, France.
| Voucher no. | Species | Providera | Collection locality | GPS coordinates |
| 0-u-4195 | Caladium lindenii (André) Madison | Chevreloup Botanical Garden | unknown | unknown |
| 2004 | X. violaceum Schott | Montpellier Botanical Garden | unknown | unknown |
| 013188 | X. atrovirens K. Koch & C. D. Bouché | Lyon Botanical Garden | unknown | unknown |
| 080119 | X. blandum Schott | Lyon Botanical Garden | Rémire-Montjoly, Cayenne, French Guyana | 4°54′35.45″N, 52°16′34.62″W |
| 070064 | X. brasiliense (Desf.) Engl. | Lyon Botanical Garden | unknown | unknown |
| 070377 | X. ceronii Croat & L. P. Hannon | Lyon Botanical Garden | Sucumbios | 00°13′06″N, 77°29′22″W |
| 031506 | X. granvillei Croat & S. A. Thomps. | Lyon Botanical Garden | Nouragues Research Center, French Guyana | 4°04′18″N, 52°43′57″W |
| 020201 | X. harlingii Croat & L. P. Hannon | Lyon Botanical Garden | Mera, Haz. San Jose, Pastaza, Ecuador | 1°27′26.15″S, 78°06′28.52″W |
| 011463 | X. hylaeae Engl. & K. Krause | Lyon Botanical Garden | Road from Tena to Puyo, Ecuador | unknown |
| 060617 | X. mexicanum Liebm. | Lyon Botanical Garden | Pena del Angel, Baja Verapaz, Guatemala | 15°05′00.74″N, 90°24′46.69″W |
| 120412 | X. piquambiense sp. nov. Croat, Scherber. & G. Ferry | Lyon Botanical Garden | Lita, Esmeraldas, Ecuador | 0°52′28.66″N, 78°28′04.07″W |
| 070769 | X. poeppigii Schott | Lyon Botanical Garden | San Juan Bosco, Morona-Santiago, Ecuador | 3°06′08.69″S, 78°31′12.39″W |
| 031520 | X. pubescens Poepp. | Lyon Botanical Garden | Perou | 11°07′49.03″S, 75°21′37.31″W |
| 060616 | X. robustum Schott | Lyon Botanical Garden | Alotenango, Sacatepequez, Guatemala | 14°29′14.15″N, 90°48′35.74″W |
| 120401 | X. sp. Croat | Lyon Botanical Garden | Colombia | unknown |
| 120452 | X. sp. nov. Croat | Lyon Botanical Garden | Medellin, Caldas, Vereda Santa Clara, Antioquia, Colombia | 5°44′40.14″N, 74°40′38.51″W |
| 031522 | X. viviparum Madison | Lyon Botanical Garden | Puyo, Pastaza, Ecuador | 1°29′32.41″S, 78°00′09.04″W |
Voucher specimens are deposited at the provider institutions.
LITERATURE CITED
- Belkhir K., Borsa P., Chikhi L., Raufaste N., Bonhomme F. 2002. GENETIX 4.04, logiciel sous Windows pour la génétique des populations. Laboratoire Génome, Populations, Intéractions CNRS UMR 5000, Université de Montpellier II, Montpellier, France.
- Billotte N., Lagoda P. J. L., Risterucci A. M., Baurens F. C. 1999. Microsatellite-enriched libraries: Applied methodology for the development of SSR markers in tropical crops. Fruits 54: 277–288 [Google Scholar]
- Chaïr H., Cornet D., Deu M., Baco M. N., Agbangla A., Duval M. F., Noyer J. L. 2010. Impact of farmer selection on yam genetic diversity. Conservation Genetics 11: 2255–2265 [Google Scholar]
- Dereeper A., Argout X., Billot C., Rami J. F., Ruiz M. 2007. SAT, a flexible and optimized Web application for SSR marker development. BMC Bioinformatics 8: 465. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kalinowski S. T., Tape M. L., Marshall T. 2007. Revising how the computer program CERVUS accommodates genotyping error increases success in paternity assignment. Molecular Ecology 16: 1099–1106 [DOI] [PubMed] [Google Scholar]
- Lebot V. 2009. Tropical root and tuber crops: Cassava, sweet potato, yams and aroids. CABI, Wallingford, United Kingdom [Google Scholar]
- Loh J. P., Kiew R., Hay A., Kee A., Gan L. H., Gan Y.-Y. 2000. Intergeneric and interspecific relationships in Araceae tribe Caladieae and development of molecular markers using amplified fragment length polymorphism (AFLP). Annals of Botany 85: 371–378 [Google Scholar]
- Offei S. K., Asante I. K., Danquah E. Y. 2004. Genetic structure of seventy cocoyam (Xanthosoma sagittifolium, Linn, Schott) accessions in Ghana based on RAPD. Hereditas 140: 123–128 [DOI] [PubMed] [Google Scholar]
- Risterucci A. M., Grivet L., N’goran J. A. K., Pieretti I., Flament M. H., Lanaud C. 2000. A high density linkage map of Theobroma cacao L. Theoretical and Applied Genetics 101: 948–955 [DOI] [PubMed] [Google Scholar]
- Schnell R. J., Goenaga R., Olano C. T. 1999. Genetic similarities among cocoyam cultivars based on randomly amplified polymorphic DNA (RAPD) analysis. Scientia Horticulturae 80: 267–276 [Google Scholar]