Genetic diversity and association mapping in a collection of selected Chinese soybean accessions based on SSR marker analysis

For broadening the narrow genetic base of modern soybean cultivars, 159 accessions were selected from the Chinese soybean collection which contained at least one of seven important agronomic traits: resistance to soybean cyst nematode (SCN) or soybean mosaic virus (SMV), tolerance to salt, cold, or drought, high seed oil content or high protein content. Genetic diversity evaluation using 55 microsatellite loci distributed across the genome indicated that a large amount of genetic diversity (0.806) and allelic variation (781) were conserved in this selected set, which captured 65.6% of the alleles present in Chinese soybean collection (1,863 accessions). On average, 9.4 rare alleles (frequency <5%) per locus were present, which were highly informative. Using model-based Bayesian clustering in STRUCTURE we distinguished four main clusters and a set of accessions with admixed ancestry. The four clusters reflected different geographic regions of origin of the accessions. Since the clusters were also clearly different with respect to the seven agronomic traits, the inferred population structure was introduced when association analysis was conducted. A total of 21 SSR markers on 16 chromosomes were identified as significantly (P < 0.01) associated with high oil content (6), high protein content (1), drought tolerance (5), SCN resistance (6) and SMV resistance (3). Twelve of these markers were located in or near previously identified quantitative trait loci (QTL). The results for both genetic relationship and trait-related markers will be useful for effective conservation and utilization of soybean germplasm.     

Features of SNP and SSR diversity in a set of ICARDA barley germplasm collection

Detection and utilization of genetic variation available in the germplasm collection for crop improvement have been the prime activities of breeders. Here a set of ICARDA barley germplasm collection comprising of 185 cultivated (Hordeum vulgare L.) and 38 wild (H. spontaneum L.) genotypes originated from 30 countries of four continents was genotyped with 68 single nucleotide polymorphism (SNP) and 45 microsatellite or simple sequence repeat (SSR) markers derived from genes (expressed sequence tags, ESTs). As two SNP markers provided 2 and 3 datapoints, a total of 71 SNPs were surveyed that yielded a total of 143 alleles. The number of SSR alleles per locus ranged from 3 to 22 with an average of 7.9 per marker. Average PIC (polymorphism information content) value for SSR and SNP markers were recorded as 0.63 and 0.38, respectively. Heterogeneity was recorded at both SNP and SSR loci in an average of 5.72 and 12.42% accessions, respectively. Genetic similarity matrices for SSR and SNP allelic data were highly correlated (r = 0.75, P < 0.005) and therefore allelic data for both markers were combined and analyzed for understanding the genetic relationships among the germplasm surveyed. Majority of clusters/subclusters were found to contain genotypes from the same geographic origins. While comparing the genetic diversity, the accessions coming from Middle East Asia and North East Asia showed more diversity as compared to that of other geographic regions. Majority of countries representing Africa, Middle East Asia, North East Asia and Arabian Peninsula included the genotypes that contained rare alleles. As expected, spontaneum accessions, as compared to vulgare accessions, showed a higher number of total alleles, higher number of alleles per locus, higher effective number of alleles and higher allelic richness and a higher number of rare alleles were observed. In summary, the examined ICARDA germplasm set showed ample natural genetic variation that can be harnessed for future breeding of barley as climate change and sustainability have become important throughout all growing areas of the world, drought/heat tolerance being the most important ones.     

Assessing genetic diversity in <Emphasis Type="Italic">Gossypium arboreum</Emphasis> L. cultivars using genomic and EST-derived microsatellites

The cultivated diploid, Gossypium arboreum L., (A genome) is an invaluable genetic resource for improving modern tetraploid cotton (G. hirsutum L. and G. barbadense L.) cultivars. The objective of this research is to select a set of informative and robust microsatellites for studying genetic relationships among accessions of geographically diverse G. arboreum cultivars. From more than 1,500 previously developed simple sequence repeat (SSR) markers, 115 genomic (BNL) and EST-derived (MUCS and MUSS) markers were used to evaluate the allelic diversity of a core panel of G. arboreum accessions. These SSR data enabled advanced genome analyses. A set of 25 SSRs were selected based both upon their high level of informativeness (PIC ≥ 0.50) and the production of clear PCR bands on agarose gels. Subsequently, 96 accessions representing a wide spectrum of diversity of G. arboreum cultivars were analyzed with these markers. The 25 SSR loci revealed 75 allelic variants (polymorphisms) ranging from 2 to 4 alleles per locus. The Neighborjoining (NJ) method, based on genetic dissimilarities, revealed that cultivars from geographically adjacent countries tend to cluster together. Outcomes of this research should be useful in decreasing redundancy of effort and in constructing a core collection of G. arboreum, important for efficient use of this genetic resource in cotton breeding.     

Diversity characterization and association analysis of agronomic traits in a Chinese peanut （Arachis hypogaea L.） mini-core collection

Association mapping is a powerful approach for exploring the molecular basis of phenotypic variations in plants. A peanut （Arachis hypogaea L.） mini-core collection in China comprising 298 accessions was genotyped using lo9 simple sequence repeat （SSR） markers, which identified 554 SSR alleles and phenotyped for 15 agronomic traits in three different environments, exhibiting abundant genetic and phenotypic diversity within the panel. A model-based structure analysis assigned all accessions to three groups. Most of the accessions had the relative kinship of less than o.05, indicating that there were no or weak relationships between accessions of the mini- core collection. For 15 agronomic traits in the peanut panel, generally the Q ＋ K model exhibited the best performance to eliminate the false associated positives compared to the Q model and the general linear model-simple model. In total, 89 SSR alleles were identified to be associated with 15 agronomic traits of three environments by the Q＋K model-based association analysis. Of these, eight alleles were repeatedly detected in two or three environments, and 15 alleles were commonly detected to be associated with multiple agronomic traits. Simple sequence repeat allelic effects confirmed significant differences between different genotypes of these repeatedly detected markers. Our results demonstrate the great potential of integrating the association analysis and marker-assisted breeding by utilizing the peanut mini-core collection.     

Development,characterization and utilization of genomic microsatellite markers in turmeric (Curcuma longa L.)

Development of a robust set of 18 genomic microsatellite markers from turmeric (Curcuma longa L.) and its effective utilization in estimating the genetic diversity of 20 turmeric accessions are described. A total of 103 alleles were detected with an average of 5.7 alleles per locus. These markers displayed varied levels of polymorphism as evident from its discriminating power ranging from 0.19 to 0.70. The UPGMA cluster analysis of genetic distance values resolved the 20 turmeric accessions into five main groups. Three sets of genetically identical accessions were detected within the analyzed accessions, suggesting a revisit of the germplasm collection strategy based on vernacular identity. The entire grouping pattern of the entities was loose and independent of their geographical origins. These polymorphic SSR markers would be useful for the population genetic studies and germplasm management of turmeric.     

Genetic structure and core collection of the World Olive Germplasm Bank of Marrakech: towards the optimised management and use of Mediterranean olive genetic resources

The conservation of cultivated plants in ex-situ collections is essential for the optimal management and use of their genetic resources. For the olive tree, two world germplasm banks (OWGB) are presently established, in Córdoba (Spain) and Marrakech (Morocco). This latter was recently founded and includes 561 accessions from 14 Mediterranean countries. Using 12 nuclear microsatellites (SSRs) and three chloroplast DNA markers, this collection was characterised to examine the structure of the genetic diversity and propose a set of olive accessions encompassing the whole Mediterranean allelic diversity range. We identified 505 SSR profiles based on a total of 210 alleles. Based on these markers, the genetic diversity was similar to that of cultivars and wild olives which were previously characterised in another study indicating that OWGB Marrakech is representative of Mediterranean olive germplasm. Using a model-based Bayesian clustering method and principal components analysis, this OWGB was structured into three main gene pools corresponding to eastern, central and western parts of the Mediterranean Basin. We proposed 10 cores of 67 accessions capturing all detected alleles and 10 cores of 58 accessions capturing the 186 alleles observed more than once. In each of the 10 cores, a set of 40 accessions was identical, whereas the remaining accessions were different, indicating the need to include complementary criteria such as phenotypic adaptive and agronomic traits. Our study generated a molecular database for the entire OWGB Marrakech that may be used to optimise a strategy for the management of olive genetic resources and their use for subsequent genetic and genomic olive breeding.     

Development of a Core Set from a Large Rice Collection using a Modified Heuristic Algorithm to Retain Maximum Diversity

A new heuristic approach was undertaken for the establishment of a core set for the diversity research of rice. As a result,107 entries were selected from the 10 368 characterized accessions. The core set derived using this new approach provideda good representation of the characterized accessions present in the entire collection. No significant differences for the mean, range, standard deviation and coefficient of variation of each trait were observed between the core and existing collections. We also compared the diversity of core sets established using this Heuristic Core Collection (HCC) approach with those of core sets established using the conventional clustering methods. This modified heuristic algorithm can also be used to select genotype data with allelic richness and reduced redundancy, and to facilitate management and use of large collections of plant genetic resources in a more efficient way.     

Diversity characterization and association analysis of agronomic traits in a Chinese peanut (Arachis hypogaea L.) mini-core collection

Association mapping is a powerful approach for exploring the molecular basis of phenotypic variations in plants.A peanut(Arachis hypogaea L.)mini-core collection in China comprising 298 accessions was genotyped using 109 simple sequence repeat(SSR)markers,which identified 554 SSR alleles and phenotyped for 15 agronomic traits in three different environments,exhibiting abundant genetic and phenotypic diversity within the panel.A model-based structure analysis assigned all accessions to three groups.Most of the accessions had the relative kinship of less than 0.05,indicating that there were no or weak relationships between accessions of the mini-core collection.For 15 agronomic traits in the peanut panel,generally the Q t K model exhibited the best performance to eliminate the false associated positives compared to the Q model and the general linear model-simple model.In total,89SSR alleles were identified to be associated with 15 agronomic traits of three environments by the Q t K model-based association analysis.Of these,eight alleles were repeatedly detected in two or three environments,and 15 alleles were commonly detected to be associated with multiple agronomic traits.Simple sequence repeat allelic effects confirmed significant differences between different genotypes of these repeatedly detected markers.Our results demonstrate the great potential of integrating the association analysis and marker-assisted breeding by utilizing the peanut mini-core collection.     

A worldwide bread wheat core collection arrayed in a 384-well plate

Bread wheat (Triticum aestivum), one of the world’s major crops, is genetically very diverse. In order to select a representative sample of the worldwide wheat diversity, 3,942 accessions originating from 73 countries were analysed with a set of 38 genomic simple sequence repeat (SSR) markers. The number of alleles at each locus ranged from 7 to 45 with an average of 23.9 alleles per locus. The 908 alleles detected were used together with passport data to select increasingly large sub-samples that maximised both the number of observed alleles at SSR loci and the number of geographical origins. A final core of 372 accessions (372CC) was selected with this M strategy. All the different geographical areas and more than 98% of the allelic diversity at the 38 polymorphic loci were represented in this core. The method used to build the core was validated, by using a second set of independent markers [44 expressed sequence tag (EST)-SSR markers] on a larger sample of 744 accessions: 96.74% of the alleles observed at these loci had already been captured in the 372CC. So maximizing the diversity with a first set of markers also maximised the diversity at a second independent set of locus. To relate the genetic structure of wheat germplasm to its geographical origins, the two sets of markers were used to compute a dissimilarity matrix between geographical groups. Current worldwide wheat diversity is clearly divided according to wheat’s European and Asian origins, whereas the diversity within each geographical group might be the result of the combined effects of adaptation of an initial germplasm to different environmental conditions and specific breeding practices. Seeds from each accession of the 372CC were multiplied and are now available to the scientific community. The genomic DNA of the 372CC, which can be entirely contained in a 384-deep-well storage plate, will be a useful tool for future studies of wheat genetic diversity. Electronic supplementary material The online version of this article (doi:) contains supplementary material, which is available to authorized users.     

A consensus list of microsatellite markers for olive genotyping

Cultivar identification is a primary concern for olive growers, breeders, and scientists. This study was aimed at examining the SSR markers retrieved from the literature and currently used in olive study, in order to select those most effective in characterizing the olive accessions and to make possible the comparison of data obtained by different laboratories. Olive microsatellite profiles were assessed by four independent laboratories, which analyzed 37 pre-selected SSR loci on a set of 21 cultivars. These SSR markers were initially tested for their reproducibility, power of discrimination and number of amplified loci/alleles. Independent segregation was tested for each pair of SSRs in a controlled cross and the allelic error rate was quantified. Some of them were finally selected as the most informative and reliable. Most of the alleles were sequenced and their sizes were determined. Profiles of the reference cultivars and a list of alleles with their sizes obtained by sequencing are reported. Several genetic parameters have been analysed on a larger set of cultivars allowing for a deeper characterization of the selected loci. Results of this study provide a list of recommended markers and protocols for olive genotyping as well as the allelic profile of a set of reference cultivars that would be useful for the establishment of a universal database of olive accessions.     

Microsatellite marker-mediated analysis of the EMBRAPA Rice Core Collection genetic diversity

The objectives of this study were to determine the genetic structure of 242 accessions from the EMBRAPA Rice Core Collection (ERiCC), to create a mini-core collection and to develop a multiplex panel of fluorescent labeled simple sequence repeats (SSRs). Eighty-six SSRs were used to identify 1,066 alleles, with an average number of 12.4 alleles/locus and average polymorphism information content (PIC)/locus of 0.75. A model-based clustering method recognized the structure of the accessions on two levels, according to their cultivation system and origin. The most divergent subgroup identified was the worldwide lowland accessions, with the highest values for gene diversity (0.75), average Rogers distance modified by Wright (0.80), average number of alleles/locus (11.7) and private alleles (132). A mini-core was assembled with the most divergent 24 lowland and upland accessions. This mini-core displayed an average distance of 0.86, an average number of alleles/locus of 8.4 and an average PIC/locus of 0.8. From the 86 SSRs, 24 were selected to compose six multiplex panels in order to optimize the process of rice genotyping. This set of markers distinguished all 242 accessions, and showed an average PIC of 0.80 and an average number of alleles/locus of 15.4, higher than the entire set of 86 SSRs. Since the heterogeneity found in lines and cultivars of ERiCC was higher than expected, it is necessary to analyze pooled DNA samples to get a better estimate of genetic variability. The SSR characterization of ERiCC clearly indicates that there is high genetic variability in rice accessions stored in genebanks worldwide which can be promptly explored by rice pre-breeding programs.     

Molecular characterization of an international cacao collection using microsatellite markers

Plant germplasm collections invariably contain varying levels of genetic redundancy, which hinders the efficient conservation and utilization of plant germplasm. Reduction of genetic redundancies is an essential step to improve the accuracy and efficiency of genebank management. The present study targeted the assessment of genetic redundancy and genetic structure in an international cacao (Theobroma cacao L.) collection maintained in Costa Rica. A total of 688 cacao accessions maintained in this collection were genotyped with 15 simple sequence repeat (SSR) loci, using a capillary electrophoresis genotyping system. The SSR markers provided a high resolution among the accessions. Thirty-six synonymously labeled sets, involving 135 accessions were identified based on the matching of multilocus SSR profiles. After the elimination of synonymous sets, the level of redundancy caused by closely related accessions in the collection was assessed using a simulated sampling scheme that compared allelic diversity in different sample sizes. The result of the simulation suggested that a random sample of 113 accessions could capture 90% of the total allelic diversity in this collection. Principal Coordinate Analysis revealed that the Trinitario hybrids from Costa Rica shared a high similarity among groups as well as among individual accessions. The analysis of the genetic structure illustrated that the within-country/within-region difference accounted for 84.6% of the total molecular variation whereas the among-country/among-region difference accounted for 15.4%. The Brazilian germplasm contributed most to this collection in terms of total alleles and private alleles. The intercountry/interregion relationship by cluster analysis largely agreed with the geographical origin of each germplasm group and supported the hypothesis that the Upper Amazon region is the center of diversity for cacao. The results of the present study indicated that the CATIE International Cacao Collection contains a high level of genetic redundancy. It should be possible to rationalize this collection by reducing redundancy and ensuring optimal representation of the genetic diversity from distinct germplasm groups. The results also demonstrated that SSR markers, together with the statistical tools for individual identification and redundancy assessment, are technically practical and sufficiently informative to assist the management of a tropical plant germplasm collection.     

SSR-based genetic diversity and structure of garlic accessions from Brazil

Garlic is a spice and a medicinal plant; hence, there is an increasing interest in ‘developing’ new varieties with different culinary properties or with high content of nutraceutical compounds. Phenotypic traits and dominant molecular markers are predominantly used to evaluate the genetic diversity of garlic clones. However, 24 SSR markers (codominant) specific for garlic are available in the literature, fostering germplasm researches. In this study, we genotyped 130 garlic accessions from Brazil and abroad using 17 polymorphic SSR markers to assess the genetic diversity and structure. This is the first attempt to evaluate a large set of accessions maintained by Brazilian institutions. A high level of redundancy was detected in the collection (50 % of the accessions represented eight haplotypes). However, non-redundant accessions presented high genetic diversity. We detected on average five alleles per locus, Shannon index of 1.2, H_O of 0.5, and H_E of 0.6. A core collection was set with 17 accessions, covering 100 % of the alleles with minimum redundancy. Overall F_ST and D values indicate a strong genetic structure within accessions. Two major groups identified by both model-based (Bayesian approach) and hierarchical clustering (UPGMA dendrogram) techniques were coherent with the classification of accessions according to maturity time (growth cycle): early-late and midseason accessions. Assessing genetic diversity and structure of garlic collections is the first step towards an efficient management and conservation of accessions in genebanks, as well as to advance future genetic studies and improvement of garlic worldwide.     

基于SSR标记的谷子遗传多样性研究

用21个分布在谷子9条染色体上的SSR标记,对120份来自于核心种质的谷子材料进行遗传多样性研究。21个标记共检查出305个等位变异,各标记检测出的等位变异数在3～26个之间,平均每个位点检测出的等位变异数为14.5个;21个位点的平均多态信息量(PIC)为0.809。基于21个SSR标记的分子鉴定,计算了120份材料间的遗传相似系数,其变化范围为0.8393～0.9672,平均值为0.8906。根据计算的遗传距离,对120份谷子材料进行UPGMA聚类,在遗传相似系数0.8865处这些材料被划分为4个类群,分类结果与这些谷子来源地生态类型总体上表现一致,分别为西北内陆类群、黄土高原内蒙古高原类群、华北平原类群以及华北平原近年育成种类群。     

Genetic structure,diversity, and allelic richness in composite collection and reference set in chickpea (<Emphasis Type="Italic">Cicer arietinum</Emphasis>L.)

Background  

Plant genetic resources (PGR) are the basic raw materials for future genetic progress and an insurance against unforeseen threats to agricultural production. An extensive characterization of PGR provides an opportunity to dissect structure, mine allelic variations, and identify diverse accessions for crop improvement. The Generation Challenge Program conceptualized the development of "composite collections" and extraction of "reference sets" from these for more efficient tapping of global crop-related genetic resources. In this study, we report the genetic structure, diversity and allelic richness in a composite collection of chickpea using SSR markers, and formation of a reference set of 300 accessions.     

Genetic diversity among cultivars, landraces and wild relatives of rice as revealed by microsatellite markers

Genetic diversity among 35 rice accessions, which included 19 landraces, 9 cultivars and 7 wild relatives, was investigated by using microsatellite (SSR) markers distributed across the rice genome. The mean number of alleles per locus was 4.86, showing 95.2% polymorphism and an average polymorphism information content of 0.707. Cluster analysis based on microsatellite allelic diversity clearly demarcated the landraces, cultivars and wild relatives into different groups. The allelic richness computed for the clusters indicated that genetic diversity was the highest among wild relatives (0.436), followed by landraces (0.356), and the lowest for cultivars. Allelic variability among the SSR markers was high enough to categorize cultivars, landraces and wild relatives of the rice germplasm, and to catalogue the genetic variability observed for future use. The results also suggested the necessity to introgress genes from landraces and wild relatives into cultivars, for cultivar improvement.     

Evaluation of Genetic Diversity and Development of a Core Collection of Wild Rice (Oryza rufipogon Griff.) Populations in China

Common wild rice (Oryza rufipogon Griff.), the progenitor of Asian cultivated rice (O. sativa L.), is endangered due to habitat loss. The objectives of this research were to evaluate the genetic diversity of wild rice species in isolated populations and to develop a core collection of representative genotypes for ex situ conservation. We collected 885 wild rice accessions from eight geographically distinct regions and transplanted these accessions in a protected conservation garden over a period of almost two decades. We evaluated these accessions for 13 morphological or phenological traits and genotyped them for 36 DNA markers evenly distributed on the 12 chromosomes. The coefficient of variation of quantitative traits was 0.56 and ranged from 0.37 to 1.06. SSR markers detected 206 different alleles with an average of 6 alleles per locus. The mean polymorphism information content (PIC) was 0.64 in all populations, indicating that the marker loci have a high level of polymorphism and genetic diversity in all populations. Phylogenetic analyses based on morphological and molecular data revealed remarkable differences in the genetic diversity of common wild rice populations. The results showed that the Zengcheng, Gaozhou, and Suixi populations possess higher levels of genetic diversity, whereas the Huilai and Boluo populations have lower levels of genetic diversity than do the other populations. Based on their genetic distance, 130 accessions were selected as a core collection that retained over 90% of the alleles at the 36 marker loci. This genetically diverse core collection will be a useful resource for genomic studies of rice and for initiatives aimed at developing rice with improved agronomic traits.     

Establishment of a Core Collection of Traditional Cuban <Emphasis Type="Italic">Theobroma cacao</Emphasis> Plants for Conservation and Utilization Purposes

Presently, Theobroma cacao L. (cacao) in Cuba is mainly cultivated in the eastern region where plantations comprise a mixture of clonal varieties, hybrids, progeny of Trinidad Selected Hybrids, and traditional—also known as ancient—cacao. The ancient genetic resources, probably the plants most closely related to the original introductions, are endangered by their progressive replacement by modern clones. To promote the conservation and utilization of these genetic resources, a representative sample of 537 traditional Cuban cacao plants was used to develop a core collection. Core collections based on 15 simple sequence repeat (SSR) markers were generated using five different sampling algorithms: random sampling, simulated annealing, stepwise clustering with random sampling, the M strategy, and maximum genetic diversity. The five core collections were designed to capture 95 % of the SSR alleles in the complete collection. The genetic, morphological, and geographical diversity of each core collection was compared with that of the entire collection. The entire collection contained 139 alleles, including 104 rare ones, with the 95 % allelic coverage threshold achieved with 133 alleles. The core collection generated by the maximum genetic diversity algorithm had the lowest number of accessions (185), the highest mean genetic distance (0.486), the lowest morphological character redundancy, and the highest diversity as assessed by the mean Shannon-Weaver diversity index (0.757). This core collection can thus serve as the basis of future improvement programs based on local genetic resources.     

High throughput analysis of grape genetic diversity as a tool for germplasm collection management

Using 20 SSR markers well scattered across the 19 grape chromosomes, we analyzed 4,370 accessions of the INRA grape repository at Vassal, mostly cultivars of Vitis vinifera subsp. sativa (3,727), but also accessions of V. vinifera subsp. sylvestris (80), interspecific hybrids (364), and rootstocks (199). The analysis revealed 2,836 SSR single profiles: 2,323 sativa cultivars, 72 wild individuals (sylvestris), 306 interspecific hybrids, and 135 rootstocks, corresponding to 2,739 different cultivars in all. A total of 524 alleles were detected, with a mean of 26.20 alleles per locus. For the 2,323 cultivars of V. vinifera, 338 alleles were detected with a mean of 16.9 alleles per locus. The mean genetic diversity (GDI) was 0.797 and the level of heterozygosity was 0.76, with broad variation from 0.20 to 1. Interspecific hybrids and rootstocks were more heterozygous and more diverse (GDI?=?0.839 and 0.865, respectively) than V. vinifera cultivars (GDI?=?0.769), Vitis vinifera subsp. sylvestris being the least divergent with GDI?=?0.708. Principal coordinates analysis distinguished the four groups. Slight clonal polymorphism was detected. The limit between clonal variation and cultivar polymorphism was set at four allelic differences out of 40. SSR markers were useful as a complementary tool to traditional ampelography for cultivar identification. Finally, a set of nine SSR markers was defined that was sufficient to distinguish 99.8% of the analyzed accessions. This set is suitable for routine characterization and will be valuable for germplasm management.     

A core collection and mini core collection of Oryza sativa L. in China

The extent of and accessibility to genetic variation in a large germplasm collection are of interest to biologists and breeders. Construction of core collections (CC) is a favored approach to efficient exploration and conservation of novel variation in genetic resources. Using 4,310 Chinese accessions of Oryza sativa L. and 36 SSR markers, we investigated the genetic variation in different sized sub-populations, the factors that affect CC size and different sampling strategies in establishing CC. Our results indicated that a mathematical model could reliably simulate the relationship between genetic variation and population size and thus predict the variation in large germplasm collections using randomly sampled populations of 700?C1,500 accessions. We recommend two principles in determining the CC size: (1) compromising between genetic variation and genetic redundancy and (2) retaining the main types of alleles. Based on the most effective scheme selected from 229 sampling schemes, we finally developed a hierarchical CC system, in which different population scales and genetic diversities allow a flexible use of genetic resources. The CC, comprising 1.7% (932) of the accessions in the basic collection, retained more than 85% of both the SSR and phenotypic variations. A mini core collection, comprising 0.3% (189) of the accessions in the basic collection, retained 70.65% of the SSR variation and 76.97% of the phenotypic variation, thus providing a rational framework for intensive surveys of natural variation in complex traits in rice genetic resources and hence utilization of variation in rice breeding.