In the last decades or so, the conventional genetic based strategies have been used successfully in QTL (quantitative trait loci) mapping for agronomically and economically important genes in different crops species such as tomato, Arabidopsis and rice, however, QTLs identified throught this method is limited to loci with large effects on the target quantitative trait variation, other techniques that enable the rapidly identification of genes playing modest roles associate with the variation of quantitative traits are also needed. Association mapping via linkage disequilibrium or LD (non-random association of alleles at different loci) offers promise in this area. The traditional approach of linkage/QTL mapping reliant on developing large mapping populations continues to suffer from lack of mapping resolution inherent in samples with limited meiotic cross-over events, while in association mapping, there may not be any need to make crosses initially to generate segregating populations, the natural variation that exists in the available germplasm can be utilized for mapping straightaway (Oraguzie et al., 2007).

LD plays a central role in association analysis, estimated of LD are important as an indication of how useful LD-base association genetics approaches can be when compared with other available mapping methods (Rafalski et al., 2004), the distance over which LD persists will determine the number and density of markers and experimental design needed to perform an association analysis. Garris et al. (2003) found out rapid decay of LD around the Xa5 locus in rice (Oryza sativa L.) happed at each 100 kb as a result of more recombination events, under this condition an average of one marker per centiMorgan (where 1 cM = 200-300 kb) were requested for a whole-genome association mapping, suggesting that candidate gene-based LD mapping could provide greater resolution than conventional QTL mapping. Candidate gene-based LD mapping was mostly advisable to be used in plants with many megabase pairs per centiMorgan but relatively rapid decay of disequilibrium over short distance such as conifer and onion (Oraguzie and Phillip, 2007). Thus, it is important to know the pattern of LD in different regions of the genome and in different populations in one organism before making an informed choice of a methodology for association genetics studies (Rafalski et al., 2004).

In an attempt towards association mapping for drought tolerance at vegetative stage in a diversity and stress adaptation rice panel, we had already examined the phenotypic performance of 184 rice germplasm accession for vegetative drought tolerance under natural drought conditions (Xiao et al., 2012a) and its population structure and genetic relatedness using 141 polymorphic SSR markers that were nearly evenly distributed at 3mb bin on the 12 rice chromosomes (Xiao et al., 2012b), with the same genotypic data, we were now trying to find out the LD model in this rice panel and to check out if a whole-genome-scanning association mapping could be used to identify loci for drought tolerance at vegetative stage. 

1 Results
Of the 156 SSR markers (Figure 1), thirteen produced monomorphic bands and two produced blurred bands, the rest 141 SSR markers produced either biallelic or polymorphic bands and were used in population structure, linkage disequilibrium and association analysis. The number of alleles per locus varied from 2 to 7, with an average of 2.73.

Figure 1 Bands produced by SSR markers. A: RM192; B: RM112; C: RM595

Significant LD was detected across the genome of the 184 rice genotypes and extent of LD varied with different chromosome (Figure 2). The pair-wise r² among the 141 SSR markers varied from 0.0000 to 0.78197, the 95^th percentile of the distribution of these estimates was 0.3886, and it was used as a population-specific threshold for this parameter as an evidence of linkage, value of r²>0.3886 were probably due to genetic linkage.

Figure 2 LD between 141 SSR markers

No genetic linkage was detected among the 141 SSR markers across the 12 chromosome, however, two pairs of SSR markers RM288 and RM464 (11.98758 Mb), RM215 and RM464 (14.61396 Mb) showed significantly higher correlation at far distance (Figure 3), all the three SSR makers were located on chromosome 9.

Figure 3 Linkage disequilibrium (r2) as a function of physical distance (Mb) among 141 markers

In cultivated barley genome-wide LD extended from 10 cm to 15 cM when evaluated with SSRs (Malysheva-otto et al., 2006) markers, and the pattern of LD was extremely population dependent.

Mather et al (2007) also reported that the extent of linkage disequilibrium differs in significant way between domesticated Asian rice and O. rufipogon, both in genome wide LD and in targeted genomic regions, higher LD was found in O. sativa variety groups. In the presence of high LD, lower marker density is required for a target region with greater potential for detecting markers strongly associated with the target gene polymorphism, even if distant physically (Agrama et al., 2007), whole-genome-scan association study is feasible in this case.

The success of association mapping efforts depends on the possibilities of separating LD due to linkage from LD due to other causes. Factors such as the mating system, the recombination rate, population structure, population history, genetic drift, directional selection, and gene fixation at different rates on different chromosome regions can all affect the patterns of LD (Gaut and Long, 2003). In general, LD decay with distance occurs at a much slower rate in self pollinated plants, such as Arabidopsis, rice, barley, durum wheat, and sorghum, than in outcrossing species. In this study, LD decayed faster than 3 MB bin, Rakshit et al. (2007) reported an LD decay of ~50Kb in indica and of ~5 Kb in O.rufipogon, which was significantly smaller as compared with 3 MB in this study, a fast decay in the genome, however, is an inference of high resolution for association analysis.

However, two pairs of markers RM288 and RM464 (11.98758 Mb), RM215 and RM464 (14.61396 Mb) all were located on chromosome 9 but at relatively far distance showed significant linkage. The same phenomenon also reported by Skot et al. (2005) in the natural population of perennial ryegrass at a genome-wide scale using AFLP genetic markers in which the majority of the linked pairs were in significant LD within genetic distance of 4.37 cM with r²=0.12, but two pairs were more than 20cM apart. These Loci might co-segregated through long term selection, as natural or artificial selection of varied trait favors co-inheritance of those loci, which was another factor cause linkage equilibrium. Selection for or against a phenotype controlled by alleles at two unlinked loci that show epistatic interaction may result in LD despite the fact that the loci are not physically linked (Palaisk et al., 2003).

3 Conclusion
Significant LD was detected across the genome of the 184 rice genotypes, selfing in O. sativa species and existing of population structure in the rice panel might be the major factors of creating high LD. However, as a high LD indicated a lower marker density would be enough to detect markers strongly associated with the trait of interest, and whole-genome-scan association study with currently used 141 SSR markers could be feasible in this case. The majority of the linked pairs were in significant LD with r²=0.3886, LD decayed faster than 3 MB bin in this mapping population, a comparatively higher resolution for association analysis with the 141 SSR markers would be expected.

4 Materials and Methods
A diversity and drought adaptation panel including 184 rice germplasm accessions which including all kinds of rice genotypes: indica parental lines such as Zhenshan 97B, Minghui 63 and IR72 etc; released indica conventional varieties such as NSIC RC9, PSB Rc18, PSB RC 82 etc; local varieties such as Azucena, Li-Jiang-Xin-Tuan-Hei-Gu etc; Aus type rice genotypes such as Dular, FR13A, Nagina 22 etc, tropical and temperate japonica type such as JAVA, Moroberekan, Cypress, deep water rice genotypes such as Aswina and Rayada; aromatic rice genotypes such as KDML 105, and interspecific rice genotypes such as WA878-6-20-1-4-P1-HB, WA880-1-32-1-1- P2-HB and WAB 891SG33, as had been reported previously (Xiao et al., 2012a) were used for DNA extraction, total genomic DNA was extracted from the leaf tissue of six seedlings per variety following the methods described by Murray and Thompson (Murray and Thompson, 1980). 156 SSR markers that were nearly evenly distributed at 3 Mb bin on the 12 rice chromosomes were selected in this study, the map position of the SSR loci was inferred on line (http://www.Gramene.org/, 2006) (Table 1). The SSR markers were first screened for their polymorphism, allele diversity of the SSR markers in 184 varieties was calculated using the PowerMarker V3.25 program (Liu and Muse, 2005), markers that produced monomorphic bands were abandoned and only those produced biallelic or polymorphic bands were used for LD analysis.

Table 1 SSR markers at each 3 Mb bin used in the experiment

Acknowledgments
This research was supported by the National Key Technologies R&D Program of China during the 12th Five-Year Plan Period (2012BAD20B03), the Innovation Fund Project of Jiangxi Academy of Agricultural Science (2012CBS013), and the Philippine Rice Research Institute (Phil Rice).

References
Agrama H.A., Eizenga G.C., and Yan W., 2007, Association Mapping of Yield and its Components in Rice Cultivars. Molecular Breeding, 19: 341-356
http://dx.doi.org/10.1007/s11032-006-9066-6 

Flavio B., and Mark E.S., 2006, Association Mapping of Kernel Size and Milling Quality in Wheat (Triticum aestivum L.) Cultivars, Genetics, 172: 1165-1177.

Garris A.J., McCouch S.R., and Kresovich S., 2003, Population Structure and its Effects on Haplotype Diversity and Linkage Disequilibrium Surrounding the Xa5 locus of Rice (Oryza sativa L.), Genetics, 165: 759-769.

Gaut B.S., and Long A.D., 2003, The lowdown on linkage disequilibrium, Plant Cell, 15: 1502-1506.
http://dx.doi.org/10.1105/tpc.150730 

Liu K.J., and Muse S.V., 2005, Power Marker: An Integrated Analysis Environment for Genetic Marker Analysis, Bioinformatics, 21: 2128-2129
http://dx.doi.org/10.1093/bioinformatics/bti282 

Mather K.A., Caicedo A.L, Polato N.R., Olsen K.M., McCouch S., and Purugganan M.D., 2007, The extent of linkage disequilibrium in rice (Oryza sativa L.), Genetics, 177(4): 2223-2232.
http://dx.doi.org/10.1534/genetics.107.079616 

Murray M.G., and Thompson W.F., Rapid isolation of high molecular weight plant DNA, Nucleic Acid Res., 1980, 8: 4321-4325
http://dx.doi.org/10.1093/nar/8.19.4321

Oraguzie N.C., Rikkerink E.H.A., and Gardiner S.E., 2007, Nihal De Silva H. Association Mapping in Plants, Springer Science + Business Media, LLC

Oraguzie N.C., Phillip L.W., Rikkerink E.H.A., and Silva H.N.D., 2007, Linkage Disequilibrium: in Association Mapping in Plants. Springer Science + Business Media, LLC

Oraguzie N.C., and Phillip L.W., 2007, An Overview of Association Mapping: in Association Mapping in Plants. Springer Science + Business Media, LLC

Palaisk K., Morgante M., Williams M., and Rafalski A., 2003, Contrasting Effects of Selection on Sequence Diversity and Linkage Disequilibrium at two Phtoene Synthease Loci, Plant Cell, 15:1795-1806
http://dx.doi.org/10.1105/tpc.012526

Rafalski A., and Morgante M., 2004, Corn and humans: recombination and linkage disequilibrium in two genomes of similar size, Trends Genet., 20(2): 103-111
http://dx.doi.org/10.1016/j.tig.2003.12.002

Rakshit S., Rakshit A., Matsumura H., Takahashi Y., and Hasegawa Y., 2007, Large-Scale DNA Polymorphism Study of Oryza sativa and O. rufipogon Reveals the Origin and Divergence of Asian Rice, Theor Appl Genet., 114:731-743
http://dx.doi.org/10.1007/s00122-006-0473-1

Skot L., Humphreys Mo, Armstead I., Heywood S., Skot K.P., Sanderson R., Thomas I.D., Sanderson R., CHorlton K.H., and Hamilton N.R.S., 2005, An association mapping approach to identify lowering time genes in natural populations of Lolium perenne (L.), Mol. Breed., 15:233-245
http://dx.doi.org/10.1007/s11032-004-4824-9

Xiao Y.L., Lei J.G., Yu C.Y., Quirino D.D.C., Jonalyn M.C., and Dindo A.T., 2012, Genetic Similarity and Population Structure in a Rice Drought Stress Adaptation Panel, Acta Agricultural Universitatis Jiangxiensis, 34(5): 886-892

Xiao Y.L., Yu C.Y., Lei J.G., Quirino D.D.C., Jonalyn M.C., and Dindo A.T., 2012, Screening of Rice Germplasm Accessions for Vegetative Drought Tolerance. Acta Agricultural Universitatis Jiangxiensis, 34(3): 0428-04