:: Korean Journal of Breeding Science ::
Journal Search Engine
Search Advanced Search Adode Reader(link)
Download PDF Export Citaion korean bibliography PMC previewer open access
ISSN : 0250-3360(Print)
ISSN : 2287-5174(Online)
Korean Journal of Breeding Science Vol.44 No.4 pp.421-432
DOI : https://doi.org/10.9787/KJBS.2012.44.4.421

Marker Development for Onion Genetic Purity Testing using SSR Finder

Chee Hark Harn1*, Hyoun-Joung Kim1, Heung-Ryul Lee1, Ji Young Hyun1, Ki Hyeon Song2, Kyu-Hyun Kim2, Jung Eun Kim3, Cheol-Goo Hur3
1Biotechnology Institute, Nongwoo Bio. Co., Ltd.,Yeoju
2Breeding Institute, Nongwoo Bio. Co., Ltd., Milyang, 3Plant Systems Engineering Research Center, Korea Research Institute of Bioscience and Biotechnology, Yuseong-gu
Received on June 15, 2012. Revised on September 18, 2012. Accepted on September 25, 2012)


Molecular genetic markers have been widely used as powerful tools for analyzing the genome. In particular, SSRmarkers have practical applications in breeding systems because they can be used in high-throughput analyses for geneticmapping, heritable diversity testing, purity analysis, and marker-assisted selection. Currently, due to technical advances in DNAsequencing, large sequence databases are available for large-scale SSR mining and marker development. Here, we describe anautomated method, the SSR Finder program, for SSR discovery in the onion sequence database, and primer design for amplifyingthe detected SSRs. A total of 1,049 SSR primers were obtained for genetic purity testing, and 100 SSR sets were analyzed in14 bulb onion breeding lines using clustering analysis. A total of 20 selected markers from screening of all 1,049 SSR primers,were finally applied for genetic purity testing in three breeding lines, NW1, NW9, and NW10. The initial tests showed that 15%,71%, and 97% of individuals within NW1, NW9, and NW10, respectively, were genetically homogeneous. These markersproduced using the SSR Finder will be useful for investigating the genetic purity of onion breeding lines.


Bulb onion (Allium cepa L.) is one of the most popular vegetables in the world (Khar et al. 2011), especially in East Asia (Song et al. 2004). Onion is a diploid (2n = 2x = 16) vegetable with a huge nuclear genome of 16,415 Mbp per 1C (Arumuganathan and Earle 1991, Kuhl et al. 2004), approximately six times larger than the genomes of maize and humans. This vegetable also has been reported to contain intrachromosomal duplications (Jones and Rees 1968), middle-repetitive sequences (Stack and Comings 1979), and Ty-1 copia-like retrotransposons (Pearce et al. 1996). Onion characteristics, such as huge genome size, biennial life cycle, and severe inbreeding depression caused by self-pollination of allogamous crops, make it more difficult to develop breeding lines and elite varieties. 

In commercial varieties and breeding lines, purity level is considered an important aspect for uniformity. For highquality seeds and vegetables, quality-control programs that monitor seeds from harvest to sale are a priority for seed companies (McDonald 1998), and morphological comparison of seedlings as a grow-out test is conducted to check the genetic purity of hybrid seeds as well as selected lines. However, these tests are time- and labor-intensive, require a large area, and sometimes produce unclear outcomes. Therefore, genetic purity tests based on molecular markers should be developed to enhance the homogeneity of genome in a breeding line. Such genetic markers would also be helpful for variety validation for the protection of breeders’ rights. 

Among several molecular marker systems, polymerase chain reaction (PCR)-based markers, including SSR (Fischer and Bachemann 2000), random amplified polymorphic DNA (RAPD; Bradeen and Havey 1995), and cleaved amplified polymorphic sequences (CAPS; van Heusden et al. 2000) are simple, fast, cheap, and easily automated (Jones et al. 1997). In particular, SSR markers have been broadly used as an ideal DNA marker for linkage map construction (Tsukazaki et al. 2008), marker-assisted selection (MAS; Kim et al. 2011), variety identification (Bredemeijer et al. 1998, Tsukazaki et al. 2010), genetic diversity studies, and genomic synteny analysis (Khar et al. 2011) because of their simplicity, and reproducibility (Jones et al. 1997). 

In the genus Allium, Fischer and Bachmann (2000) first reported the development of SSR markers from a bulb onion genomic library. Kuhl et al. (2004) detected 336 SSR cores among 313 expressed sequence tags (EST) and developed 88 EST-SSR markers. In addition, Martin et al. (2005) located 35 EST-SSRs, 43 single nucleotide polymorphisms (SNPs), and 4 insertion-deletion (InDel) markers on a medium-density linkage map in bulb onion. In bunching onion, 1,940 SSR clones were isolated from genomic libraries (Song et al. 2004, Tsukazaki et al. 2007), and Tsukazaki et al. (2008) constructed a linkage map based on SSR markers from both bunching and bulb onions, and assigned eight basic chromosomes of Allium. cepa. In addition, SSR markers were used in heterogeneity assays and variety traceability tests in bunching onion (Tsukazaki et al. 2006). McCallum et al. (2008) also reported onion genetic diversity tested using 56 EST-SSR markers and four genomic SSR markers. 

To develop SSR markers, several steps are required; these include collecting sequence information, identifying SSRs, parsing SSRs, and designing and ordering primers. Several programs such as BatchPrimer3 (You et al. 2008) and Web-Sat (Martins et al. 2009) are generally used for each step and are manually linked among programs from output to input data to produce an appropriate format for each program used. Because output data files are not compatible with all programs, further work on format conversion is required, which is very tedious and time-consuming. Recent technological innovations and intense research have led to the rapid accumulation of large amounts of sequence information. Yet few easy-to-use program-linked steps for sequence analysis, specific sequence discovery, and primer design for amplification of core sequences have been developed. Hence, to make better use of these data and simultaneously reduce time and labor, SSR Finder program was developed (unpublished). 

Here, we report the development of SSR markers of onion using the SSR Finder. Twenty SSR markers deduced from SSR Finder were successfully used in 14 breeding lines of bulb onion and applied to sib-crossed progeny of three breeding lines to check for purity. The results demonstrate that the SSR markers that are present here can effectively help breeders assess the genetic purity and diversity of onion. 


Plant materials and genomic DNA extraction

Bulb onion seeds of NW1 to NW14 (Table 1, Fig. 1), which had been sib-crossed over 5–7 generations for onion breeding, were used for the SSR marker screening. 

Table 1. Description on origin and traits of 14 onion breeding lines.

Fig. 1. Dendrogram obtained from UPGMA cluster analysis using horticultural traits of the 14 breeding lines.

Germinated young leaves were ground using a Retsch MM301 ball mill (Daigger, USA) for 5 min, and genomic DNA (gDNA) was prepared following the method Murray and Thompson (1980). The concentration of gDNA was measured with a micro-volume spectrophotometer, ASP-2680 (Avans Biotechnology, Taiwan). 

SSR primer design

A total of 33,461 ESTs and 11,540 gDNA sequences of onion (Table 2) were collected from the National Center for Biotechnology Information (NCBI; http://www.ncbi.nlm.nih. gov/) and the Computational Biology and Functional Genomics Laboratory (http://compbio.dfci.harvard.edu/). To find SSRs with two to five motifs over a length of 12 bp, Sputnik (http:// espressosoftware.com/sputnik/index.html) and RepeatMasker (http://www.repeatmasker.org/cgi-bin/WEBRepeatMasker) software were used (Fig. 2). To obtain PCR products of 100 to 350 bp in length containing SSRs, primers were designed using Primer3 (http://frodo.wi.mit.edu/primer3/) with lengths of 20 to 24 bp and a melting temperature of 60℃. Each primer was designed to amplify each SSR, but some primer amplicons contained two SSRs because of the close (less than 50 bp) location between two SSR regions. Primer sets were synthesized by the Bioneer Corporation (Korea). 

Table 2. Number of gDNA and EST sequences used for SSR primer design and SSR marker selection using the RepeatMasker and Sputnik programs.

Fig. 2. Diagram of the SSR Finder program. Each box represents a step for SSR primer design. The conditions and methods for each step are given in each box.

SSR Finder and primer sets

To develop SSR markers, the SSR Finder program was developed with four steps (Fig. 2): FASTA-formatted data submission as a starting point, SSR detection and parsing in subsequent steps, and finally primer design as a wellordered output file. These were linked into one process on a Linux system. 

To start the program with sequence data submission, a public database was used and divided into ESTs and gDNA sources. A total of 45,001 onion sequences consisting of 33,461 EST and 11,540 gDNA sequences were analyzed using SSR Finder (Table 2). SSR primer sets of 670 and 379 were designed from ESTs and gDNA, respectively. The designed SSR primers were 3.3% gDNA and 2% EST sequences. 

To assess whether the SSR primers amplify PCR products and polymorphic bands, the first 100 sets from a total 1,049 designed SSRs were randomly selected and applied to 14 breeding lines (Fig. 3). The selected primer set of 50 (eSSR1–eSSR50) was deduced from the EST sequences, and a further set (gSSR1–gSSR50) was deduced from gDNA (Table 3). After screening of all 1,049 primers, selected 20 SSR markers (Table 3) were adjusted for purity testing of three onion lines. 

Fig. 3. Band pattern amplified with the SSR primer sets gSSR6 (A), gSSR16 (B), gSSR43 (C), and eSSR27 (D) in 14 onion breeding lines. Arabic numeral is the line number. The arrow indicates a polymorphic band.

PCR conditions and agarose gel electrophoresis

PCR reactions were conducted in a total volume of 20 μl containing 10 × buffer, 0.2 mM dNTP, 0.5 mM forward and reverse primers (Tables 3 and 4), 40 ng genomic DNA, and 0.5 U Taq DNA polymerase (Genet Bio Inc., Korea). PCR was performed on a GeneAmp PCR System 9700 (Applied Biosystems, USA) with the followed conditions: denaturation at 94℃ for 2 min; 35 cycles of 94℃ for 30 s, 60℃ for 30 s, and 72℃ for 30 s; and final elongation at 72℃ for 10 min. PCR products were separated in EtBrstained 3% agarose gels at 200 V for 90 min, their polymorphism was scored with a UV transilluminator, and the polymorphic information content (PIC) was calculated according to Anderson’s formula (Anderson et al. 1993). 

Table 3. Information on gSSR and eSSR primer sets.

Table 3.Continued.

Table 3.Continued.

Table.3. Continued.

Table 4. Information on SSR primer sets used for genetic purity tests.

Clustering analysis

All band patterns of 32 SSR markers obtained from the 14 onion breeding lines were scored and used for clustering analysis. Their genetic distances were calculated using TREECON (van de Peer and de Wachter 1994) according to the developers’ instructions, using the unweighted pairgroup method with arithmetic averaging (UPGMA) and clustering analysis of the genetic similarity coefficient matrices (Nei and Li 1979). 


SSR detection

Together with designed primer sets, information on detected SSR regions was also provided as FASTA-type output files. For searching SSR regions, two common methods, RepeatMasker and Sputnik, were used (Table 2). Both methods detected similar numbers of SSR regions, about half of which were method-specific. That is, among 50 eSSRs, 44 and 41 were detected with RepeatMasker and Sputnik, respectively; 35 were found with both methods Nine and six eSSRs were found only with RepeatMasker and Sputnik, respectively. Among gSSRs, 24 were detected with both programs, and 9 and 17 were found with RepeatMasker- and Sputnik-specific SSRs, respectively. Among the selected 100 SSR primer sets (Table 3), 59 were found with both programs, but 18 and 23 were RepeatMasker- and Sputnik-specific SSR regions, respectively. 

A total of 32 of 100 primers produced polymorphisms in the 14 lines were selected as elite sets for genetic diversity and purity tests based on the PIC value. Among 32 sets, 15 and 17 originated from EST and gDNA sequences, respectively. A total of 7 and 8 SSRs were specifically detected with either RepeatMasker or Sputnik, respectively, and the other 17 sets were found with both programs. Three eSSRs and four gSSRs were RepeatMasker-specific, and eight gSSRs were Sputnik-specific. 

Repeat length and polymorphism

In the case of SSR motif length (Fig. 4), tri-nucleotides and more than hexa-nucleotides were most common in eSSRs, but mono-nucleotide was poorly detected. Four tetra-nucleotides and one hexa-nucleotide were used for SSR primer design. Among these, three primer sets originated from tetra-nucleotides and the hexa-nucleotide, respectively, produced polymorphic bands. In gSSRs, four from 10 tri-nucleotides, five from 13 tetra-nucleotides, one hexanucleotide, respectively, amplified polymorphisms. 

Fig. 4. The number of SSR primer sets (A) and polymorphic SSR sets (B) based on the SSR motif length used among 14 onion breeding lines.

Among the SSRs, all except four sets were well amplified, with less than 1 kb size (Fig. 3). eSSR sets mainly produced mono bands while gSSRs produced poly bands. Amplicon sizes from eSSRs were mainly longer than those expected, due to added intron regions. Polymorphic bands amplified from 100 sets were converted into PIC values, which ranged between 0.138 and 0.997 (Table 3). A total of 17 gSSRs and 15 eSSRs produced polymorphisms among the 14 lines, with average PIC values of 0.775 and 0.697, respectively. These polymorphic bands were dominant in most cases. 

Phenetic relationship and purity test

Polymorphisms of 44 alleles that were produced by 32 primers were used for genetic clustering of the 14 NW lines (Fig. 5). In the clustering, breeding lines were shown to have useful genetic backgrounds, with broad distances for diverse variety development, but no relationship with bulb shape, bulb color, or origin was found. 

Fig. 5. Dendrogram of the 14 breeding lines using UPGMA cluster analysis of genetic distance based on 44 alleles from 32 SSR markers. The number on the right side is the line number.

After screening 1,049 primer combinations in the 14 NW lines, specific bands for separating breeding lines were selected. From these, 20 elite primers were used for purity testing of three lines, NW1, NW9, and NW10 (Table 4). A total of 9, 9, and 10 primer combinations were examined in 75 progeny of NW1, 77 of NW9, and 115 of NW10. Within progeny of NW1, NW9 and NW10, three, one, and one combinations, respectively, produced polymorphic bands (Table 5). From these results, individuals of 64 NW1, 22 NW9, and 3 NW10 had different band patterns, and these progenies were eradicated to improve homogeneity within these lines. The other progenies of 11 NW1, 55 NW9, and 112 NW10 were selected for further screening. Among three lines, NW1 was the most heterogeneous and NW10 was the most homogeneous, based on the detected alleles using SSRs. 

Table 5. Number of primer sets and progeny used in the purity test.


SSR Finder

SSR Finder has the advantage of fast and easy processing, and could be helpful to analyze massive sequence databases and obtain SSR primer sets. This software supports all steps from SSR detection to SSR primer design with just one click, and provides useful information, such as accession number of input sequence, SSR types and location, motif sequence, annealing temperature of primers, and expected product size. Moreover, because all treated input and output files in SSR Finder were in FASTA format, which is compatible with most commonly used programs as an easily downloaded sequence type from public databases, the program can be useful for researchers even when they do not have computational information (Rudd et al. 2003). 

To find an SSR region using SSR Finder, two methods, RepeatMasker and Sputnik, were assessed (Fig. 2). Both found half of the detected SSRs, but the other SSRs (41%) were program-specific (Table 2). Therefore, to detect SSRs, SSR Finder operated using these two methods can provide more SSR regions than a Web-based program. Further study to more friendly and free used program in window system, has been progressed. 

Repeat length and type

Kuhl et al. (2004) found that di-nucleotide and tri-nucleotide repeats represented 35% and 60% of 336 EST-SSRs, respectively, with a frequency of 1 SSR/25 kb and an average repeat length of 7.3, using the MISA program (Thiel et al. 2003), but no strong associations between repeat type or length and rates of polymorphism were detected. In accordance with this result, the present study detected tri-nucleotide repeats in total SSRs and eSSRs (Fig. 4), but no major relationship with repeat type or length or rates of polymorphism. Also, in gSSRs, similar amounts of di-nucleotide to penta-nucleotide repeat types were found (Table 3). Among 12 (AT)n, 5 SSRs amplified polymorphic bands, but a repeat motif type did not show any relationship with polymorphism level. 

Fischer and Bachman (2000) reported that GT repeats were the majority in 30 developed SSR markers in bulb onion. Song et al. (2004) sequenced 94 genomic clones, and revealed that among 52 SSRs, the GT di-nucleotide motif was more frequent than GA (49 GT motifs and 1 GA motif) in bunching onion; in contrast, GA is more frequent than GT in most plant species. In bunching onion, Tsukazaki et al. (2007) isolated 1796 SSRs from gDNA libraries, which contained 74% (GT)n repeats (n > 5) and 17.5% (GA)n repeats. They amplified an average of 2.28 alleles with average PIC values of 0.739 from 100 SSR loci. In the present study (Table 3), 13 di-nucleotide repeat SSRs and 12 (AT)n repeats were detected. Most di-nucleotide repeats originated from gDNA sequences and the (AT)n repeat was dominant. This result might be related to the different onion lines and agarose gel system used for SSR screening. Used agarose gel system is not able to detect small size differences less than 10 bp, which can lose most reported SSRs with small differences and co-dominant inheritance, but nevertheless the gel system is more 3 times cheaper than acrylamide gel system. Therefore, agarose gel system has been effectively adapted for mass-screening in crop breeding (Lee et al. 2007, Lee et al. 2009). Moreover, these SSR markers under agarose gel system could be informative and newly applied for diversity tests and MAS in onion breeding, because the developed SSR markers might originate from different repeat regions than those previously reported. 

Phenetic relationships and purity tests

Fischer and Bachmann (2000) and McCallum et al. (2008) reported that genetic clustering in onion germplasm was related to geographical origin. However, Yang et al. (2001) did not find any relationship between clustering and bulb shape, color, or origin, which is accordant with the result in Figure 4, and this might be related to the narrow origin of the sources. 

For purity testing of NW1, NW9, and NW10, 20 sets that allowed line differentiation from the other 13 lines were examined (Table 4). Within their sib-crossed progeny, there were line-specific bands that proved the heterozygosity level of the genomic background as detected with the 20 SSR markers. A total of 15% of NW1, 71% of NW9, and 97% of NW10 progeny were selected to improve homogeneity within these lines (Table 5). On seven occasions, the sib-crossed NW10 progeny indicated very high homogeneity from the purity test.

Bunching onion varieties exhibit a high degree of genetic heterogeneity (Haishima et al. 1993, Tsukazaki et al. 2006). However, Rouamba et al. (2001) documented narrow genetic diversity of inbred varieties of onion (A. cepa) based on nine enzyme system and onion sequences, although withinpopulation diversity was higher than between-population diversity. Kuhl et al. (2004) also reported a restricted genetic background using onion sequence analysis. From the effects of these phenomena on onion, there was a low level of polymorphism but a high level of PIC. In the present study, line-specific bands and high PIC levels were also found. Moreover, Song et al. (2004) and Tsukazaki et al. (2006) suggested an SSR-tagged breeding scheme using a small number of highly polymorphic SSR loci, based on the degree of uniformity at the selected SSR loci.

Therefore, these SSR markers with high PIC levels will probably be useful for collecting progeny with high genetic homogeneity for onion breeding, although these sets require further study to understand whether they are well segregated on the whole genome, and to obtain representative marker sets for genetic tests. Also, the SSR Finder program and the developed SSR markers could be sequenceinformative for genetic diversity and purity testing in crops, but few markers have yet been developed. 


1.Anderson JA, Churchill GA, Autrique JE, Tanksley SD, Sorrells ME. 1993. Optimizing parental selection for genetic linkage maps. Genome 36: 181-186.
2.Arumuganathan K, Earle ED. 1991. Nuclear DNA content of some important plant species. Plant Mol. Biol. Rep. 9: 208-218.
3.Bradeen JM, Havey MJ. 1995. Randomly amplified polymorphic DNA in bulb onion and its use to assess inbred integrity. J. Am. Soc. Hort. Sci. 120: 752-758.
4.Bredemeijer GMM, Arens P, Wouters D, Visser D, Vosman B. 1998. The use of semi-automated fluorescent microsatellite analysis for tomato cultivar identification. Theor. Appl. Genet. 97: 584-590.
5.Fischer D, Bachmann K. 2000. Onion microsatellite for germplasm analysis and their use in assessing intra- and inter-specific relatedness within the subgenus Rhizirideum. Theor. Appl. Genet. 101: 153-164.
6.Haishima M, Kato J, Ikehashi H. 1993. Isozyme polymorphism in native cultivars of Japanese bunching onion (Allium fistulosum L.). Japan J. Breed. 43: 537-547.
7.van Heusden AW, Shigyo M, Tashiro Y, Ginkel RV, Kik C. 2000. AFLP linkage group assignment to the chromosomes of Allium cepa L. via monosomic addition lines. Theor. Appl. Genet. 100: 480-486.
8.Jones CJ, Edwards KJ, Castaglione S, Winfield MO, F Sala, van de Wiel C, Bredemeijer B, Vosman B, Matthes M, Daly A, Brettschneider R, Bettini P, Buiatti M, Maestri E, Malcevschi A, Marmiroli N, Aert R, Volckaert G, Rudea J, Linacero R, Vazquez A, Karp A. 1997. Eproducibility testing of RAPD, AFLP and SSR markers in plants by a network of European laboratories. Mol. Breed. 3: 381-390.
9.Jones R, Rees H. 1968. Nuclear DNA variation in Allium. Heredity 23: 591-605.
10.Khar A, Lawande KE, Negi KS. 2011. Microsatellite marker based analysis of genetic diversity in short day tropical Indian onion and cross amplification in related Alliums pp. Genet. Resour. Crop Evol. 58: 741-752.
11.Kim K, Lee H, Hyun JY, Won D, Hong DO, Cho H, Lee KA, Her N H, Lee JH, Harn CH. 2011. Application of disease resistance markers for developing elite tomato varieties and lines. Kor. J. Hort. Sci. & Tech. 29: 336-344.
12.Kuhl JC, Cheung F, Yuan Q, Martin W, Zewdie Y, MaCallum J, Catanach A, Rutherford P, Sink KC, Jenderk M, Prince JP, Town CD, Havey MJ. 2004. A unique set of 11,008 onion expressed sequence tags reveals expressed sequences and genomic differences between the monocot orders Asparagales and Poales. Plant Cell 16: 114-125.
13.Lee JH, Yeo US, Kwak DY, Park DS, Song YC, Kim CS, Jeon MG, Shin MS, Oh BG, Yang SJ, Nam MH, Ku YC, Kim JK, Sohn JK. 2007. Utilization of RM28493 for marker-assisted selection to brown planthopper resistance (Bph1) in rice (Oryza sativa L.). Korean J. Breed. Sci. 39: 406-411.
14.Lee SK, Moon JK, Jeong KH, Seo MJ, Kim YH, Yun HT, Kim JK. 2009. A new variety "Singang" with resistance to soybean mosaic virus by molecular marker assisted selection. Korean J. Breed. Sci. 41: 568-573.
15.Martins WS, Cesar D, Lucas S, Neves KFS, Bertioli DJ. 2009. WebSat - A web software for microsatellite marker development. Bioinformation 3(6): 282-283.
16.Martin WJ, McCallum J, Shigyo M, Jakse J, Kuhl JC, Yamane N, Pither-Joyce M, Gökce AF, Sink KC, Town CD, Havey MJ. 2005. Genetic mapping of expressed sequences in onion and in silico comparison with rice show scant colinearity. Mol. Genet. Genomics 274: 197-204.
17.McCallum J, Thomson S, Pither-Joyce M, Kene F. 2008. Genetic diversity analysis and single-nucleotide polymorphism marker development in cultivated bulb onion based on expressed sequence tag-simple sequence repeat markers. J. Amer. Soc. Hort. Sci. 133: 810-818.
18.McDonald MB. 1998. Seed quality assessment. Seed. Sci. Res. 8: 265-275.
19.Murray MG, Thompson WF. 1980. Rapid isolation of highmolecular-weight plant DNA. Nucleic Acids Res. 8: 4321-4326.
20.Nei M, Li WH. 1979. Mathematical model for studying genetic variation in terms of restriction endonucleases. Proc. Natl. Acad. Sci. USA 76: 5269-5273.
21.Pearce SR, Pich U, Harrison G, Flavell AJ, Heslop-Harrison JS, Schubert I, Kumar A. 1996. The Ty1-copia group retrotransposons of Allium cepa are distributed throughtout the chromosomes but are enriched in the terminal heterochromatin. Chromosome Res. 4: 357-364.
22.van de Peer Y, de Wachter R. 1994. TREECON for Windows: a software package for the construction and drawing of evolutionary trees for the Microsoft Windows environment. Comput. Applic. Biosci. 10: 569-570.
23.Rouamba A, Sandmeier M, Sarr A, Ricroch A. 2001. Allozyme variation within and among populations of onion (Allium cepa L.) from West Africa. Theor. Appli. Genet. 103: 855-861.
24.Rudd S, Mewes H, Mayer KFX. 2003. Sputnik: a database platform for comparative plant genomics. Nucleic Acids Research 31: 128-132.
25.Song YS, Suwabe K, Wako T, Ohara T, Nunome T, Kojima A. 2004. Development of microsatellite markers in bunching onion (Allium fistulosum L.). Breed. Sci. 54: 361-365.
26.Stack SM, Comings DE. 1979. The chromosomes and DNA of Allium cepa. Chromosoma. 70: 161-181.
27.Thiel T, Michalek W, Varshney RK, Graner A. 2003. Exploiting EST databases for the development and characterization of gene-derived SSR-markers in barley (Hordeum vulgare L.). Theor. Appl. Genet. 106: 411-422.
28.Tsukazaki H, Fukuoka H, Song Y, Yamashita K, Wako T, Kojima A. 2006. Considerable heterogeneity in commercial F1 varieties of bunching onion (Allium fistulosum) and proposal of breeding scheme for conferring variety traceability using SSR markers. Breed. Sci. 56: 321-326.
29.Tsukazaki H, Honjo M, Yamashita K, Ohara T, Kojima A, Ohsawa R, Wako T. 2010. Classification and identification of bunching onion (Allium fistulosum) varieties based on SSR markers. Breed. Sci. 60:139-152.
30.Tsukazaki H, Nunome Y, Fukuoka H, Kanamori H, Kono I, Yamashita K, Wako T, Kojima A. 2007. Isolation of 1,796 SSR clones from SSR-enriched DNA libraries of bunching onion (Allium fistulosum). Euphytica 157: 83-94.
31.Tsukazaki H, Yamashita K, Yaguchi S, Masuzaki S, Fukuoka H, Yonemaru J, Kanamori H, Kono I, Hang TTM, Shigyo M, Kojima A, Wako T. 2008. Construction of SSR-based chromosome map in bunching onion (Allium fistulosum). Theor. Appl. Genet. 117: 1213-1223.
32.Yang BK, Kim DH, Kim I, Yi YB, Suh JK, Nam J, Jeong SJ. 2001. Analysis of genetic diversity of onion germplasm using RAPD. J. Kor. Soc. Hort. Sci. 42: 533-539.
33.You FM, Huo N, Gu YQ, Luo M, Ma Y, Hane D, Lazo GR, Dvorak J, Anderson OD. 2008. BatchPrimer3: a high throughput web application for PCR and sequencing primer designi. BMC Bioinformatics. 9: 253.