• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of geneticsGeneticsCurrent IssueInformation for AuthorsEditorial BoardSubscribeSubmit a Manuscript
Genetics. Aug 2008; 179(4): 2239–2252.
PMCID: PMC2516094

Quantitative Trait Loci (QTL) Analysis For Rice Grain Width and Fine Mapping of an Identified QTL Allele gw-5 in a Recombination Hotspot Region on Chromosome 5


Rice grain width and shape play a crucial role in determining grain quality and yield. The genetic basis of rice grain width was dissected into six additive quantitative trait loci (QTL) and 11 pairs of epistatic QTL using an F7 recombinant inbred line (RIL) population derived from a single cross between Asominori (japonica) and IR24 (indica). QTL by environment interactions were evaluated in four environments. Chromosome segment substitution lines (CSSLs) harboring the six additive effect QTL were used to evaluate gene action across eight environments. A major, stable QTL, qGW-5, consistently decreased rice grain width in both the Asominori/IR24 RIL and CSSL populations with the genetic background Asominori. By investigating the distorted segregation of phenotypic values of rice grain width and genotypes of molecular markers in BC4F2 and BC4F3 populations, qGW-5 was dissected into a single recessive gene, gw-5, which controlled both grain width and length–width ratio. gw-5 was narrowed down to a 49.7-kb genomic region with high recombination frequencies on chromosome 5 using 6781 BC4F2 individuals and 10 newly developed simple sequence repeat markers. Our results provide a basis for map-based cloning of the gw-5 gene and for marker-aided gene/QTL pyramiding in rice quality breeding.

RICE (Oryza sativa L.) is the world's most important cereal crop and is considered a model cereal crop due to its relatively small genome size, vast germplasm collection, enormous repertoire of molecular genetic resources, and efficient transformation system (Paterson et al. 2005). Rice grain quality consists of several components: cooking texture, palatability, flavor, grain appearance, milling efficiency, and nutritional quality. Among these, the cooking, eating, and appearance qualities constitute important economic concerns that influence rice production in many rice-producing areas of the world. Rice grain appearance is mainly specified by grain shape as defined by grain length, width and the length–width ratio (LWR), and chalkiness of the endosperm (Zhang 2007). Preference for rice grain shape, however, varies among consumer groups. For instance, long and slender grain varieties are preferred by consumers in the United States and Western Europe and in most Asian countries or areas, including China, India, Pakistan, and Thailand; in contrast, consumers in Japan, South Korea, and Sri Lanka prefer short and bold grain cultivars (Unnevehr et al. 1992; Juliano and Villareal 1993). Therefore, breeding for the appropriate grain shape needs to be considered in the context of market preference. Additionally, rice grain width and shape can greatly affect other important rice quality traits, such as endosperm chalkiness, milling efficiency, eating, and cooking properties (McKenzie and Rutger 1983; Takita 1989; Chauhan et al. 1995; Tan et al. 2000; Xu et al. 2004). Similarly, rice grain width and shape also play important roles in determining grain yield (Lin and Wu 2003; Yoon et al. 2006). Notably, grain shape is also an important indicator of the evolution of crops due to the continuous selection for large seeds during early domestication. However, small seeds are usually favored by natural selection, as small seed size is frequently associated with more seeds per plant, early maturity, and wider geographic distribution (Harlan 1992). Therefore, the great economic importance associated with grain width and shape necessitates in-depth study of their genetic basis and development mechanism(s) to better understand biological development processes and to facilitate breeding in rice.

Several independent studies have identified a number of QTL for rice grain width in primary mapping populations (Lin et al. 1995; Huang et al. 1997; Redoña and Mackill 1998; Tan et al. 2000; Xu et al. 2002; Aluko et al. 2004; Li et al. 2004b). In these studies, grain width was evaluated in a single environment, so the stability of the resultant QTL remained unknown. However, this stability information is critical in determining the usefulness of the QTL in marker-aided selection (MAS) breeding. Additionally, among the 31 grain-width QTL detected in these studies, one was consistently identified in the same genomic region of rice chromosome 5, with an average percentage of phenotypic variation explained (PVE) of 35.9%. Whether this QTL corresponds to only one gene or to a cluster of genes, each with relatively small contribution to the total genetic effects, remains unknown. Thus, it is necessary to establish whether this QTL is a single Mendelian factor and to further narrow down its location on genetic and physical maps.

Primary mapping populations such as F2:3, backcross (BC), doubled haploids (DH), and recombinant inbred lines (RILs) are not appropriate for fine mapping of an identified QTL, as they simultaneously segregate at the whole parental chromosome segments (Yamamoto et al. 2000). On the other hand, chromosome segment substitution lines (CSSLs) or near-isogenic lines can simplify or eliminate the effect of genetic background on the expression of QTL and therefore facilitate fine mapping and isolation of the alleles of target QTL (Frary et al. 2000; Fridman et al. 2004; He et al. 2006; Konishi et al. 2006; Wan et al. 2006). A recent study reported that QTL GW2 for rice grain width and weight on chromosome 2 encodes a RING-type E3 ubiquitin ligase that negatively regulates cell division by targeting its substrate(s) to proteasomes for proteolysis. Loss of GW2 results in increasing cell numbers and wider spikelet hull, thus enhancing rice grain width, weight, and yield (Song et al. 2007). Furthermore, a major grain-width QTL qGW-5 on chromosome 5 was consistently detected in all eight environments including two cropping seasons (2001 and 2002) and four locations (Nanjing, Jinhu, Donghai, and Hainan in China), and the donor IR24 allele of qGW-5 had a significant and stable effect on decreasing rice grain width in the isogenic background of Asominori (Wan et al. 2005).

The objectives of this study were the following: (1) to obtain information on rice grain-width QTL including additive QTL, epistatic QTL, QTL by environment interactions, and gene action of identified additive QTL; (2) to dissect the major QTL qGW-5 into a single gene; and (3) to localize the gw-5 gene to a narrow genomic region.


Plant materials:

Four different populations (RILs, CSSLs, BC4F2, and BC4F3) were used to conduct QTL analysis and fine mapping in this study. Seventy-one RILs were derived from the single cross between Asominori and IR24 by single-seed descent (Tsunematsu et al. 1996). A total of 66 CSSLs with a genetic background of mostly Asominori, denoted as CSSL1-66, were produced by crossing and backcrossing 19 selected RILs with Asominori (without selection) until the BC3F1 generation, as described in supplemental Figure S1 (Kubo et al. 1999; Wan et al. 2004a). Of two narrow-grain lines (CSSL28 and 29) harboring the gw-5 gene, CSSL28 was used to develop BC4F2 and BC4F3 populations by backcrossing to Asominori with subsequent self-pollination.

Field experiment design and phenotypic evaluation:

This study was carried out in 12 environments, E1–E12, including five locations and eight cropping seasons (supplemental Table 1S). Asominori, IR24, and their 71 RILs were grown in 4 environments (E1–E4), and the parental varieties and their 66 CSSLs were planted in 8 environments (E1–E8). Each experimental plot consisted of two replicates, with the design of 10 rows × 10 plants for each line. At maturity, each plot was harvested in bulk. After drying, grains were milled using the method of Wan et al. (2004a), and grain width was determined on the basis of the method of Wan et al. (2005).

Two thousand one hundred seventy-one, 1248, 2465, and 897 CSSL28/Asominori F2 (BC4F2) plants were grown in E9, E10, E11, and E12, respectively. One hundred BC4F3 families (16 plants per family) were planted in E11. At maturity, seeds were collected from primary panicles. Paddy and brown rice were used to evaluate grain length and grain width using the methods of Wan et al. (2005, 2006), and the LWR, which represents the grain shape, was calculated as the ratio of grain length/width. Of the 6781 BC4F2 plants, 805 homozygous individuals with wide grain were used to precisely map the gw-5 gene.

DNA preparation, PCR protocol, and DNA marker analysis:

DNA was extracted from fresh rice leaves of BC4F2 plants as described by Dellaporta et al. (1983). PCR was performed using the procedure of Chen et al. (1997). PCR products were separated on an 8% nondenaturing polyacrylamide gel and detected using the silver staining method of Sanguinetti et al. (1994). The required density of markers in the genomic region harboring the gw-5 gene was achieved using previously published simple sequence repeat (SSR) and expressed sequence tag (EST) markers on rice chromosome 5 (McCouch et al. 2002; Wu et al. 2002) as well as the SSR markers newly developed in this study. New SSR markers were designed using the method of Wan et al. (2006). Primer sequences, map positions, and amplified lengths of 10 new SSR markers are listed in supplemental Table 2S.

QTL analysis and gene action of identified additive QTL:

The linkage map of the Asominori/IR24 RIL population including 375 markers (Tsunematsu et al. 1996) was used for QTL analysis. Tests of QTL additive effects, epistatic interactions and QTL × environment interactions (QEI) were performed using the QTL Mapper 1.0 program (Wang et al. 1999). The LOD = 4.03 (up to P = 0.005) was used as the threshold for determining the presence of additive or epistatic QTL. The relative contribution was calculated as the PVE by the QTL. Additionally, gene action of identified additive QTL was evaluated by a t-test to show the presence of significant differences between the phenotypic values of the recurrent parent Asominori and those of CSSLs harboring the target QTL alleles derived from the donor parent IR24.

Fine mapping of the gw-5 gene:

For fine mapping of the gw-5 gene, the bulked-extreme and recessive-class approach as described in Zhang et al. (1994) was used to calculate recombination frequencies between the gw-5 gene and molecular markers in the 805 homozygous BC4F2 plants with wide grain. Thus, the recombination frequency = (N1 + N2/2)/N, where N is the total number of wide-grain plants surveyed, N1 is the number of wide-grain plants with the band pattern of the narrow-grain parent, and N2 is the number of wide-grain plants with heterozygous band patterns.


Stability of QTL for rice grain width and gene action of identified QTL:

The phenotypic distributions of rice grain width in the RIL populations grown in the E1–E4 environments (supplemental Table 1S) are shown in Figure 1. Variance among genotypes (G) was highly significant for grain width, but was not significant among environments (E). The significant G × E interaction explained 3.5% of the total phenotypic variation. Six additive effect QTL for rice grain width were identified and mapped to six chromosomes, with LOD values ranging from 7.68 to 23.77 (Table 1). Among the QTL, qGW-5 was consistently located in the Y1060L–R569 interval on chromosome 5 in populations grown across all four environments, and the average PVE by qGW-5 was 24.3%. The IR24 allele at the qGW-5 locus was found to reduce grain width by an average of 0.115 mm. Moreover, stability of qGW-5 was relatively high, as its QEI was not significant. Additionally, 11 pairs of epistatic QTL for rice grain width were detected and mapped on 10 chromosomes. Of these, three digenic interactions occurred between an additive effect QTL (qGW-1, qGW-4, or qGW-12) and a modifying factor (Table 2).

Figure 1.
Phenotypic distributions of rice grain width in the Asominori/IR24 recombinant inbred lines in four environments (E1–E4).
The putative additive QTLs and their environmental interactions for grain width of milled rice
The putative epistatic QTLs and their environmental interactions for grain width of the milled rice

Eighteen lines were selected from the total 66 CSSLs and used to analyze gene action of the six additive effect QTL. The donor IR24 segments harboring qGW-5 and qGW-9 were transferred into 2 (CSSL28 and 29) and 4 CSSLs (CSSL53–55 and 57), respectively (Kubo et al. 1999). Significant difference in grain width of milled rice was observed between Asominori and each of CSSL28 and CSSL29 across all eight environments, indicating that the gene action of qGW-5 was significant and stable in the Asominori genetic background (Figure 2). With qGW-9, significant difference in grain width between Asominori and each of the four target CSSLs was found only in a few environments. The direction of the effect of qGW-9 in CSSL53 and CSSL55 was consistent with that in the RIL population (Table 1), but opposite to that in CSSL54 (Figure 2). Thus, the gene action of qGW-9 was sensitive to both the genetic background and environmental conditions. Similar results were also observed for qGW-1, qGW-4, qGW-10, and qGW-12 (supplemental Figure S2). Therefore, the observed stability and gene action of these QTL indicates that qGW-5 is the most important genetic factor that controls rice grain-width difference between the Asominori and IR24 parental lines.

Figure 2.
Grain width of the milled rice in Asominori and target CSSLs carrying qGW-5 or qGW-9 alleles in eight environments (E1–E8). * and ** indicate significance levels of 5 and 1%, respectively; ns, nonsignificant difference. ...

QTL qGW-5 controlled both rice grain width and length–width ratio in the BC4F2 population:

In E9, the phenotypic distributions of grain width of paddy and brown rice in the CSSL28/Asominori F2 population appeared to be bimodal with boundaries of 3.28 and 2.80 mm, respectively (Figure 3, A and B). The ratio of narrow- (1921) to wide-grain individuals (250) was 7.68:1, which does not fit with the expected segregation ratio of 3:1 for the inheritance of a single gene. Similarly, the ratio of small to large LWR individuals was 1:7.68 (Figure 3, E and F). Interestingly, all 250 small LWR individuals showed wide grains, and all 1921 large LWR plants presented narrow grains. Moreover, significantly negative correlations (r = −0.94** and −0.94**) were observed between grain width and LWR of paddy rice as well as brown rice in the total 2171 BC4F2 individuals. Furthermore, although LWR is a complex trait reflecting grain length and width, QTL for grain length had relatively small effects on the phenotypic variation of LWR in the BC4F2 population, as shown by Figure 3, C and D, which clearly indicated that minor factor(s) conferred the phenotypic variations of grain length. These results suggest that qGW-5 mainly affected both rice grain width and LWR in the BC4F2 population. This is further supported by the fact that the ratios of narrow- to wide-grain individuals were 7.05:1, 7.36:1, and 7.54:1 in the CSSL28/Asominori F2 populations that included 1248 plants in E10, 2465 in E11, and 897 in E12 (Figure 4, A–C); their corresponding ratios of small to large LWR individuals were 1:7.05, 1:7.36, and 1:7.54, respectively.

Figure 3.
Frequency distributions of grain width, grain length, and length–width ratio in the CSSL28/Asominori BC4F2 population (2171 individuals) in E9. (A) Grain width of paddy rice; (B) grain width of brown rice; (C) grain length of paddy rice; (D) grain ...
Figure 4.
Frequency distributions of grain width of paddy rice in the CSSL28/Asominori F2 populations grown in E10, E11, and E12. (A) Grain width of paddy rice in E10 (1248 individuals); (B) grain width of paddy rice in E11 (2465 individuals); (C) grain width of ...

Dissecting QTL qGW-5 into a single gene, gw-5:

To dissect qGW-5 into a single gene, we analyzed four genetic models that could possibly lead to the observed distortion of segregation in the BC4F2 population. Wide rice grain could be controlled by one of the following models: (1) two recessive linked genes at the qGW-5 locus, (2) a recessive gene linked with gamete sterility gene(s), (3) a recessive gene linked with gene(s) for weak germination or seedling viability, or (4) a recessive gene linked with hybrid partial sterility gene(s). If case 1 is correct, genotypes of the markers near qGW-5 including Asominori(A)/A, A/CSSL28(L), and L/L should show the normal 1:2:1 ratio in the BC4F2 population. However, the ratios of 8 SSR markers in the interval 20.6–39.2cM near qGW-5 did not exhibit the predicted 1:2:1 ratio (Figure 5, A1–D1). If case 2 is the correct model, the above ratios of 7.68:1, 7.05:1, 7.36:1, and 7.54:1 should correspond to 1:3.9:3.8, 1:3.7:3.4, 1:3.7:3.6, and 1:3.9:3.8 for A/A:A/L:L/L, respectively. But the observed actual ratios listed in Figure 5, A1–D1, clearly exclude case 2. Additionally, rice seeds with the genotype of A/A, A/L, or L/L had similar germination ability and seedling viability (supplemental Table 3), ruling out the model presented in case 3.

Figure 5.
The ratios of three genotypes of SSR markers and frequency distributions of grain width in 300 BC4F2 individuals across four environments (E9–E12). A1–D1 represent the ratios of three genotypes of 12 SSR markers in E9–E12, respectively; ...

Several observations support the fourth case. First, complete fertility was observed for Asominori, IR24, and CSSL28, whereas partial sterility occurred in Asominori/CSSL28 F1 plants, which evidently resulted from hybrid partial sterility gene(s). Second, an indicajaponica hybrid partial sterility gene S31(t) mapped to 20.9–24.7 cM on chromosome 5 in our laboratory (Zhao et al. 2006) was linked with qGW-5 located at 36.4–44.9 cM (Table 1). Third, eight SSR markers in the 20.6–39.2cM region showed regularly progressive segregation distortions in the three genotypes, with ratios of LL + AL/AA from 7.33:1 at 39.2 cM to 14.79:1 at 20.6 cM (Figure 5, A1), indicating that the closer the distance from the SSR markers to the S31(t) gene, the stronger the degree of segregation distortion. Thus, the S31(t) gene played a crucial role in inducing the distortion of the ratios. In contrast, four SSR markers (RM163, RM430, RM506, and RM408) far from the qGW-5 locus showed a nearly normal ratio of 3:1 for LL + AL/AA (Figure 5, A1). Similar results were also observed in E10–E12 (Figure 5, B1–D1). Additionally, when 300 randomly selected BC4F2 plants were divided into three groups on the basis of the three genotypes of the RMw513 marker, 35, 37, 42, and 36 plants with the genotype A/A in E9–E12, respectively, showed wide rice grains (Figure 5, A2–D2), whereas all other narrow-grain plants presented the genotype of A/L or L/L, indicating that the distorted segregation of marker genotypes was consistent with that of grain-width phenotype. Finally, all individuals in 35 BC4F3 families with the genotype A/A at the RMw513 locus showed wide grains without segregation. Of 2408 BC4F3 plants derived from 152 BC4F2 plants with the genotype A/L, the ratio of narrow- (2124) to wide-grain plants (284) was 7.48:1, consistent with that in the BC4F2 population (Figures 3 and and4).4). These findings clearly indicate that qGW-5 behaves like a single gene in the CSSL28/Asominori F2 population and that both wide rice grains and small length–width ratio are mainly controlled by a recessive qGW-5 allele, named gw-5.

Environmental impact on the effect of the gw-5 gene:

Since wide rice grain was controlled by the gw-5 gene in the BC4F2 population, the environmental impact on the effect of the gw-5 gene could be evaluated by comparing the average grain width of BC4F2 individuals with the genotype of gw-5/gw-5 among the E9–E12 environments. The largest effect of the gw-5 gene occurred in E9 and E11, with average grain widths of 3.36 ± 0.05 and 3.36 ± 0.04 mm, and 2.88 ± 0.04 and 2.88 ± 0.05 mm for paddy rice and brown rice, respectively; gw-5 had the smallest effect in E10, in which paddy rice and brown rice had average grain widths of 3.20 ± 0.05 and 2.66 ± 0.04 mm, respectively. Moreover, significant differences of grain width were observed between E9/E11 and E12, as well as E12 and E10 (Table 3), thus showing that the environment had a significant impact on the effect of the gw-5 gene. Likewise, a similar environmental impact was also found on the effect of the GW-5 gene (supplemental Table 4).

Comparison of grain width in the BC4F2 individuals with wide-grain phenotype (gw-5/gw-5) across four environments (E9–E12)

Fine mapping of the gw-5 gene in a recombination hotspot region on chromosome 5:

For fine mapping, 805 wide-grain homozygotes of the total 6781 BC4F2 individuals were used to calculate the recombination frequency between the gw-5 gene and the surrounding molecular markers. Using previously published SSR and EST markers, we detected three SSR markers (RM3328, RM3322, and RM5874) and one EST marker (C53703) on one side of the gw-5 gene, and one SSR marker (RM5994) on the other side (Figure 6C). Ten polymorphic SSR markers were then designed using DNA sequences of seven BAC/PAC contigs in Nipponbare (supplemental Table 2S and Figure 6B). Among the markers, RMw530 and RMw513 had 23 and 6 recombinants with the gw-5 gene in the 805 homozygotes, respectively (Figure 6C). Thus, the gw-5 gene was located in the 1.8-cM genetic region on chromosome 5. Searching for the contigs OJ1097_A12 (harboring RMw530) and B1007D10 (carrying RMw513) in the Nipponbare genome (http://www.gramene.org), we found that RMw530 and RMw513 reside at 5,309,078 and 5,358,806 bp on chromosome 5, respectively. Thus, the gw-5 gene was narrowed down to a 49.7-kb genomic region (Figure 6C).

Figure 6.
Genetic and physical maps of the gw-5 gene on rice chromosome 5. (A) QTL mapping of the qGW-5 locus. (B) Six BAC/PAC contigs around the gw-5 locus. (C) Fine mapping of gw-5 using newly developed SSR markers. The red superscript a indicates that there ...

The 17.1-cM genetic region (RM5874–RM5994) can be divided into 10 intervals, with ratios of physical-to-genetic distance of 226.3, 261.7, 183.9, 68.9, 27.6, 40.0, 14.2, 257.3, 267.4, and 255.4 kb/cM (Figure 7). The gw-5 gene was located in the RMw530–RMw513 interval with the kb/cM of 27.6. In rice, the genomewide average is estimated at 244 kb/cM (Chen et al. 2002), indicating that the crossover frequency between RMw530 and RMw513 was approximately nine times that of the whole genome in rice.

Figure 7.
Physical-to-genetic distance ratios in 10 marker intervals near the gw-5 gene.

On the basis of the available sequence annotation, we found five predicted candidate genes in the 49.7-kb region (http://www.ncbi.nlm.nih.gov/BLAST; http://www.softberry.com). Of these genes, one had unknown function, and the other four encoded an auxin-responsive protein IAA16, a 20S proteasome β-subunit, a hexokinase 6, and the Ulp1 protease-like protein. This result will be useful in the molecular cloning and functional characterization of the gw-5 gene.


Dissection of the QTL qGW-5 into a recessive gene gw-5 despite the nontraditional Mendelian segregation ratio:

In this study, QTL qGW-5 was separated as a recessive gene, gw-5, and narrowed down to a 49.7-kb genomic region, though the segregation ratios did not fit with the inheritance mode of a single Mendelian factor under multiple environments (Figures 3–6).). Several observations support this conclusion:

  1. The ratios of narrow- to wide-grain individuals were 7.68:1, 7.05:1, 7.36:1, and 7.54:1 in the CSSL28/Asominori BC4F2 population in four environments (E9–E12) (Figures 3 and and4),4), indicating that the environment likely had little effect on the distorted segregation ratios.
  2. A similar distorted segregation ratio (i.e., 7.48:1) was also observed in the BC4F3 population, excluding the possibility that the distorted segregation could be attributed to the partial false F1 hybrid seeds of CSSL28/Asominori.
  3. Eight SSR markers in the 20.6–39.2 cM region on rice chromosome 5 showed progressive distortions of segregation in three genotypes, with ratios of (GW-5/GW-5 + GW-5/gw-5)/gw-5/gw-5 from 6.14 at 39.2 cM to 14.79 at 20.6 cM (Figure 5, A1–D1), showing that the closer the SSR markers to the hybrid partial-sterility gene S31(t) (Zhao et al. 2006), the stronger the degree of segregation distortion. However, four SSR markers located far from qGW-5 and the S31(t) gene showed a nearly normal ratio of 3:1 (Figure 5, A1–D1).
  4. Complete fertility was observed for Asominori, IR24 and CSSL28, whereas partial sterility occurred in Asominori/CSSL28 F1 plants, which evidently resulted from the hybrid partial sterility. Thus, the S31(t) gene should play a crucial role in triggering the distorted ratios of marker genotypes near the gw-5 gene.
  5. Among 300 BC4F2 plants, 35, 37, 42, and 36 plants with the gw-5/gw-5 genotype in E9–E12, respectively, showed wide rice grains (Figure 5, A2–D2), whereas all other narrow-grain plants presented the genotype of GW5/gw-5 or GW-5/GW-5. This indicated that the distorted segregation of marker genotypes was consistent with that of phenotypic values of grain width.
  6. Rice seeds with the genotype of GW-5/GW-5, GW-5/gw-5, or gw-5/gw-5 had similar germination ability and seedling viability (supplemental Table 3), which ruled out the possibility that a gene(s) for weak germination or seedling viability linked to the gw-5 gene resulted in the distorted segregation in the BC4F2 population.
  7. We used 805 wide-grain plants and 10 SSR markers to narrow down the gw-5 gene to a 49.7-kb genomic region on rice chromosome 5 (Figure 6), further demonstrating that qGW-5 corresponded to the recessive gw-5 gene. Therefore, although it did not show the inheritance mode of a single gene in the BC4F2 population, qGW-5 was separated as the gw-5 gene, and the observed distorted segregation might be caused by its linkage to the hybrid partial-sterility S31(t) gene located at a genetic distance of ~15.0 cM (Figure 6; Zhao et al. 2006).

Dissection of the genetic modes of QTL underlying complex quantitative traits:

The QTL qGW-5 allele, gw-5, was narrowed down to a 49.7-kb genomic region exhibiting recombination hotspots (Figures 3–7),), illustrating that a QTL could be treated as a single Mendelian factor to perform research on functional genomics of a quantitative trait. Using a strategy of forward genetics, many QTL have been dissected into single genes for fine mapping and map-based cloning in recent years. The genetic modes of QTL established as single Mendelian factors can be categorized into at least six different types (Figure 8, A–F), i.e., the allele of a QTL corresponding to (1) a dominant major gene for the phenotypic distribution with a discrete boundary and a ratio of 3:1 (e.g., in rice, Fan et al. 2006; Wan et al. 2006; and in wheat, Cuthbert et al. 2006) or (2) with a partially overlapping boundary and a ratio of 3:1 in the secondary F2 population (e.g., in rice, Wan et al. 2004b; Yamamoto et al. 2000); (3) an additive major gene for the phenotypic distribution with two discrete boundaries and a ratio of 1:2:1 (e.g., in soybean, Yamanaka et al. 2005) or (4) with two partially overlapping boundaries and a ratio of 1:2:1 (e.g., in rice, Yamamoto et al. 1998); (5) an additive major gene for the continuous distribution in a secondary F2 population and a ratio of 1:2:1 in F3 families (e.g., in rice, Monna et al. 2002; Takeuchi et al. 2003; and in wheat, Röder et al. 2008) or (6) a non-Mendelian segregating major gene linked to other gene(s) for gamete sterility, hybrid partial sterility, weak germination, or seedling viability (this study). These modes not only disclose the complex genetic properties of QTL, but also provide some useful rules of how to interpret the results with non-Mendelian segregation ratios in advanced backcross progenies when establishing QTL into single genes and fine mapping QTL. Since the first use of a complete RFLP linkage map to identify QTL controlling fruit mass, concentration of soluble solids, and fruit pH in the tomato by Paterson et al. (1988), >1000 QTL regulating agronomic and economic traits in crops have been identified, but only a few have been precisely mapped and isolated. One important reason may be the complex genetics of QTL as described above. Therefore, greater research contribution by the geneticists in the field of QTL mapping is required so that more QTL will be dissected into individual genes on the basis of the genetic modes of different QTL (Figure 8) and fine mapping and map-based cloning carried out. In addition, overlapping substitute lines have been used to achieve high resolution mapping of QTL in plants (Paterson et al. 1990; Eshed and Zamir 1995). Recently, a reverse genetics approach was employed for fine mapping of QTL, in which recombinants in a target region harboring a QTL of interest were identified by comprehensive genotyping followed by phenotyping of the resultant informative recombinants (Li et al. 2004a). By combining this method with tests of advanced backcross progenies, several QTL were precisely localized in narrow genomic regions (Chen and Tanksley 2004; Thomson et al. 2006). Furthermore, QTL isogenic recombinant analysis has been developed to speed up fine mapping of a QTL with a single population (Peleman et al. 2005). The progress on QTL fine mapping would facilitate research on the functional genomics of complex quantitative traits and provide new insights into the genetic theory of quantitative traits.

Figure 8.
Six models for dissecting a QTL into a single gene in a secondary F2 population. (A) The allele of a QTL belongs to a dominance major gene leading to a ratio of 3:1 with a discrete boundary. (B) The allele of a QTL is a dominant major gene resulting in ...

The significance of this work to the basic understanding of genetic principles of QTL across taxa in plants:

Classical quantitative genetics describes the aggregate behavior of suites of genes influencing a quantitative trait. However, an understanding of quantitative inheritance at the molecular level requires detailed descriptions on genetic properties of individual genes controlling the quantitative trait, which is made possible by QTL mapping (Paterson 1995). Basic properties of individual QTL, such as additivity, dominance, and overdominance, have been analyzed, and a wide range of different modes of gene action is evident for various QTL (Paterson et al. 1991; Redoña and Mackill 1996). Many basic genetic phenomena are reconciled with the results from QTL mapping, such as transgression, heterosis, the number of genes for a quantitative trait, and the importance of epistasis (Paterson 1995; Yano and Sasaki 1997). However, several key genetic questions from most identified QTL in plants still remain unclear, e.g., the number of genes at individual QTL, the kinds of phenotypic distributions controlled by individual QTL alleles, and the types of genetic effects of individual QTL alleles. It is relatively difficult to answer these questions by primary QTL mapping strategies on the basis of F2, RIL, DH, and BC populations, while CSSLs or NILs harboring target QTL are useful for in-depth exploration of genetic modes and effects of QTL in plants. In this study, we used BC4F2 and BC4F3 populations to identify the QTL qGW-5 corresponding to a single gene gw-5 with a recessive genetic effect. However, the phenotypic segregation ratio controlled by the gw-5 gene did not fit the 3:1 of a single-gene inheritance (Figures 3–6).). On the basis of this finding, we summarized at least six different modes of how to dissect a QTL into a single gene, which can be applied in plants such as rice, soybean, and wheat. These modes not only enable researchers in the field of QTL mapping to dissect more QTL into individual genes, but also clearly indicate the types of genetic effects of QTL alleles on the basis of phenotypic distributions in secondary F2 and F3 populations (Figure 8). More importantly, this study provides a classic example of the dissection of genetic properties and modes of identified QTL and further performance of QTL fine mapping when distorted phenotypic distributions are observed in advanced backcross progenies. Therefore, this report describes significant findings for dissecting and furthering the understanding of genetic principles of QTL. Furthermore, the research strategies used in this study are applicable among other taxa in plants.

Additionally, cereal species such as rice (Oryza), maize (Zea), sorghum (Sorghum), and wheat (Triticum) began to diverge from a common ancestor only ~50–70 million years ago and have undergone largely parallel selection regimes associated with domestication and improvement over the past several thousand years (Paterson et al. 2003). This makes it possible to perform comparative mapping among these cereal species. As a result, a close correspondence among QTL affecting complex traits such as seed size has been demonstrated in sorghum, sugarcane, maize, wheat, barley, and rice (Paterson et al. 1995). Thus, QTL analysis in one taxon may provide significant clues to similar findings in other taxa, and correspondence of QTL across diverse taxa also provides strong empirical support for the use of model systems in research of complex phenotypes in plants (Paterson 1995). On the other hand, the rice genome sequence provides a platform for comparing and integrating genomic information about diverse cereals and together with genetic maps and sequence samples from other cereals yields new insights into both the shared and independent dimensions of cereal evolution (Paterson et al. 2003). Therefore, the results of genetics and QTL fine mapping of rice seed dimensions obtained here are applicable toward the molecular dissection of evolution and domestication among cereal species.

A potential application of the gw-5 gene for improving grain quality of rice on the basis of molecular design breeding strategy:

Biological data (including QTL/gene mapping) continue to dramatically accumulate with the rapid development of biotechnology and genomics. Nevertheless, the lack of appropriate tools and methods of simulation breeding renders it difficult to integrate this information into traditional crop breeding. Strategies such as molecular design breeding (MDB) can resolve, at least partly, this conflict and facilitate crop breeding. Before going to the field, breeders design a blueprint for obtaining particular breeding objectives, by which the crossing and selection process can be simulated and optimized, thus greatly enhancing breeding efficiency (Wang et al. 2007a,b). Until now, most of the successful research on MDB or MAS breeding has been conducted on resistance to diseases or insects conferred by major genes, but few efforts have addressed the improvement of complex quantitative traits, possibly due to the lack of information on QTL epistasis, QEI effects, gene action of QTL, and markers closely linked to target QTL (Li 2001). On the other hand, the continuous phenotypic distribution in progeny and measurement of rice grain width and shape after the reproductive stage make it difficult for breeders to efficiently improve grain appearance using conventional selection methods. In addition, rice grain width and shape can greatly affect other important quality traits such as endosperm chalkiness, milling efficiency, cooking texture, and palatability of cooked rice, and play an important role in determining rice grain yield. Therefore, grain width and shape are good candidate traits for MDB in rice breeding programs.

In the meantime, qGW-5 can be recommended as a desirable QTL for improving rice grain quality due to the following observations:

  1. qGW-5 has been shown to control rice grain shape not only in Indica/Japonica (Xu et al. 2002; Tan et al. 2005; Wan et al. 2005, 2006) and Indica/Indica (Lin et al. 1995; Tan et al. 2000; Lin and Wu 2003) populations, but also in a cultivated rice germplasm collection based on the whole-genome association mapping strategy (Iwata et al. 2007). The rice germplasm set consists of 281 landraces and 51 modern cultivars originating from 23 rice-producing countries and is considered extremely useful for rice improvement in breeding experiments (Kojima et al. 2005). Therefore, qGW-5 should be applicable and important in the improvement of grain appearance quality in traditional and molecular rice breeding programs.
  2. The expression stability of qGW-5 is relatively high both at the level of multiple QTL (Tables 1 and and2)2) and an individual QTL allele across multiple environments (Figure 2).
  3. qGW-5 accounts for a large proportion of the overall phenotypic variation (Table 1; Wan et al. 2005).
  4. Dissecting qGW-5 into the gw-5 gene as well as its fine mapping make it a desirable candidate for MDB, as the deleterious gene(s) near the gw-5 allele can be easily removed by MAS, and 11 flanking markers within a 2.0-cM distance (Figure 6C) can be readily used by breeders for MAS of the gw-5 into elite cultivars.

Additionally, other major QTL for grain length (Wan et al. 2006), endosperm chalkiness (Wan et al. 2004b), starch viscosity (rapid viscosity analyzer profile characteristics), amylose content, and protein content (Wan et al. 2004a; Weng et al. 2006; our unpublished data) in rice have also been characterized in detail, dissected into single genes, and precisely mapped. The MDB methodology and computer programs have recently been developed and employed by our laboratory (Wang et al. 2004, 2007a,b; Li et al. 2007). Thus, QTL/gene pyramiding on the basis of MDB and MAS strategies should be feasible for simultaneously improving multiple rice quality traits. In this regard, the field experiment work is ongoing.

Localization of the gw-5 gene in the recombination hotspot region:

Recombination at the meiotic stage plays an important role in genome evolution, crop cross-breeding, and map-based cloning of genes/QTL. In this study, four recombination hotspot regions (Figure 7) corresponded to three BAC/PAC contigs (OJ1725_E07, OJ1097_A12, and B1007D10) in Nipponbare (Figure 6B). We found that the G + C contents of OJ1725_E07, OJ1097_A12, and B1007D10 contigs were 40.3, 41.4, and 42.5%, the densities of the predicted genes were 13.5, 11.3, and 16.3 (genes/100 kb), and the percentages of repetitive sequences were 9.8, 10.2, and 8.4%, respectively. Thus, the relatively high densities of the predicted genes and low repetitive sequences were found to be comparable to those in the hotspot regions of rice chromosome 1 (Wu et al. 2003) and in the genomes of Arabidopsis and other cereals (Arabidopsis Genome Initiative 2000; Ware and Stein 2003). The G + C content in these hotspot regions was almost the same as in the hotspot regions of rice chromosome 1 (Wu et al. 2003), but relatively lower than that in hotspot regions of Arabidopsis, yeast, and human (Jeffreys et al. 1998; Gerton et al. 2000), indicating that a specific mechanism may exist for activating recombination in the hotspots of rice chromosomes 1 and 5. This novel finding creates an opportunity, as well as poses a challenge, for molecular geneticists in the further exploration of the relationship between fine structure and crossover frequency of rice chromosomes by dissecting the mechanism of high recombination frequencies in the four regions surrounding the gw-5 gene between the two parents (Asominori and CSSL28). Similarly, linkage mapping strategies have been used to investigate the recombination hotspots on chromosomes of a wide variety of organisms, e.g., wheat (Faris et al. 2000), neurospora (Yeadon et al. 2004), mouse (Kelmenson et al. 2005), and human (Minin et al. 2007).


We greatly thank A. Yoshimura, Kyushu University, Japan, for kindly providing us with the RIL and CSSL populations and genotype data. We also thank San Francisco Edit (http://www.sfedit.net) for kindly proofreading, editing, and refining the English of this manuscript. This research was supported by a grant from the National High Technology Research and Development Program of China (2003AA222131) and the project 111 program.


  • Aluko, G., C. Martinez, J. Tohme, C. Castano, C. Bergman et al., 2004. QTL mapping of grain quality traits from the interspecific cross Oryza sativa × O. glaberrima. Theor. Appl. Genet. 109 630–639. [PubMed]
  • Arabidopsis Genome Initiative, 2000. Analysis of the genome sequence of the flowering plant Arabidopsis thaliana. Nature 408 796–815. [PubMed]
  • Chauhan, J. S., V. S. Chauhan and S. B. Lodh, 1995. Comparative analysis of variability and correlations between quality components in traditional rainfed upland and lowland rice. Indian J. Genet. Plant Breed. 55 6–12.
  • Chen, K. Y., and S. D. Tanksley, 2004. High-resolution mapping and functional analysis of se2.1: a major stigma exsertion quantitative trait locus associated with the evolution from allogamy to autogamy in the genus Lycopersicon. Genetics 168 1563–1573. [PMC free article] [PubMed]
  • Chen, M. S., G. Presting, W. B. Barbazuk, J. L. Goicoechea, B. Blackmon et al., 2002. An integrated physical and genetic map of the rice genome. Plant Cell 14 537–545. [PMC free article] [PubMed]
  • Chen, X., S. Temnykh, Y. Xu, Y. G. Cho and S. R. McCouch, 1997. Development of a microsatellite framework map providing genome-wide coverage in rice (Oryza sativa L.). Theor. Appl. Genet. 95 553–567.
  • Cuthbert, P. A., D. J. Somers, J. Thomas, S. Cloutier and A. Brule-Babel, 2006. Fine mapping Fhb1, a major gene controlling fusarium head blight resistance in bread wheat (Triticum aestivum L.). Theor. Appl. Genet. 112 1465–1472. [PubMed]
  • Dellaporta, S. L., J. Wood and J. B. Hicks, 1983. A plant DNA mini preparation: Version II. Plant Mol. Biol. Rep. 1 19–21.
  • Eshed, Y., and D. Zamir, 1995. An introgression line population of Lycopersicon pennellii in the cultivated tomato enables the identification and fine mapping of yield-associated QTL. Genetics 141 1147–1162. [PMC free article] [PubMed]
  • Fan, C. C., Y. Z. Xing, H. L. Mao, T. T. Lu, B. Han et al., 2006. GS3, a major QTL for grain length and weight and minor QTL for grain width and thickness in rice, encodes a putative transmembrane protein. Theor. Appl. Genet. 112 1164–1171. [PubMed]
  • Faris, J. D., K. M. Haen and B. S. Gill, 2000. Saturation mapping of a gene-rich recombination hot spot region in wheat. Genetics 154 823–835. [PMC free article] [PubMed]
  • Frary, An., T. C. Nesbitt, Am. Frary, S. Grandillo, E. V. D. Knaap et al., 2000. fw2.2: A quantitative trait locus key to the evolution of tomato fruit size. Science 289 85–88. [PubMed]
  • Fridman, E., F. Carrari, Y. S. Liu, A. R. Fernie and D. Zamir, 2004. Zooming in on a quantitative trait for tomato yield using interspecific introgressions. Science 305 1786–1789. [PubMed]
  • Gerton, J. L., J. Derisi, R. Shroff, M. Lichten, P. O. Brown et al., 2000. Global mapping of meiotic recombination hotspots and coldspots in the yeast Saccharomyces cerevisiae. Proc. Natl. Acad. Sci. USA 97 11383–11390. [PMC free article] [PubMed]
  • Harlan. J., 1992. Crops & Man.American Society of Agronomy/Crop Science Society of American, Madison, WI.
  • He, G. M., X. J. Luo, F. Tian, K. G. Li, Z. F. Zhu et al., 2006. Haplotype variation in structure and expression of a gene cluster associated with a quantitative trait locus for improved yield in rice. Genome Res. 16 618–626. [PMC free article] [PubMed]
  • Huang, N., A. Parco, T. Mew, G. Magpantay, S. McCouch et al., 1997. RFLP mapping of isozymes, RAPD and QTLs for grain shape, brown planthopper resistance in a doubled haploid rice population. Mol. Breed. 3 105–113.
  • Iwata, H., Y. Uga, Y. Yoshioka, K. Ebana and T. Hayashi, 2007. Bayesian association mapping of multiple quantitative trait loci and its application to the analysis of genetic variation among Oryza sativa L. germplasms. Theor. Appl. Genet. 114 1437–1449. [PubMed]
  • Jeffreys, A. J., J. Murray and R. Neumann, 1998. High-resolution mapping of crossovers in human sperm defines a minisatellite-associated recombination hotspot. Mol. Cell 2 267–273. [PubMed]
  • Juliano, B. O., and C. P. Villareal, 1993. Grain Quality Evaluation of World Rices.International Rice Research Institute, Manila, The Philippines.
  • Kelmenson, P. M., P. Petkov, X. Wang, D. C. Higgins, B. J. Paigen et al., 2005. A torrid zone on mouse chromosome 1 containing a cluster of recombinational hotspots. Genetics 169 833–841. [PMC free article] [PubMed]
  • Kojima, Y., K. Ebana, S. Fukuoka, T. Nagamine and M. Kawase, 2005. Development of an RFLP-based rice diversity research set of germplasm. Breed. Sci. 55 431–440.
  • Kubo, T., K. Nakamura and A. Yoshimura, 1999. Development of a series of Indica chromosome segment substitution lines in Japonica background of rice. Rice Genet. Newsl. 16 104–106.
  • Li, Z. K., 2001. QTL mapping in rice: a few critical considerations, pp. 153–171 in Rice Genetics IV, edited by G. S. Khush, D. S. Brar and B. Hardy. Science Publishers, New Delhi.
  • Li, H., G. Ye and J. Wang, 2007. A modified algorithm for the improvement of composite interval mapping. Genetics 175 361–374. [PMC free article] [PubMed]
  • Li, J. M., M. Thomson and S. R. McCouch, 2004. a Fine mapping of a grain-weight quantitative trait locus in the pericentromeric region of rice chromosome 3. Genetics 168 2187–2195. [PMC free article] [PubMed]
  • Li, J. M., J. H. Xiao, S. Grandillo, L. Y. Jiang, Y. Z. Wan et al., 2004. b QTL detection for rice grain quality traits using an interspecific backcross population derived from cultivated Asian (O. Sativa L.) and African (O. glaberrium S.) rice. Genome 47 697–704. [PubMed]
  • Lin, H. X., S. K. Min, Z. M. Xiong, H. R. Qian, J. Y. Zhuang et al., 1995. RFLP mapping of QTLs for grain shape traits in indica rice (Oryza sativa L. subsp. Indica). Scientia Agric. Sin. 28 1–7.
  • Lin, L. H., and W. R. Wu, 2003. Mapping of QTLs underlying grain shape and grain weight in rice. Mol. Plant Breed. 1 337–342.
  • McCouch, S. R., L. Teytelman, Y. B. Xu, K. B. Lobos, K. Clare et al., 2002. Development and mapping of 2240 new SSR markers for rice (Oryza sativa L.). DNA Res. 9 199–207. [PubMed]
  • McKenzie, K. S., and J. N. Rutger, 1983. Rice grain shape and rice grain quality. Crop Sci. 23 306–313.
  • Minin, V. N., K. S. Dorman, F. Fang and M. A. Suchard, 2007. Phylogenetic mapping of recombination hotspots in human immunodeficiency virus via spatially smoothed change-point processes. Genetics 175 1773–1785. [PMC free article] [PubMed]
  • Monna, L., X. Lin, S. Kojima, T. Sasaki and M. Yano, 2002. Genetic dissection of a genomic region for a quantitative trait locus, Hd3, into two loci, Hd3a and Hd3b, controlling heading date in rice. Theor. Appl. Genet. 104 772–778. [PubMed]
  • Paterson, A. H., 1995. Molecular dissection of quantitative traits: progress and prospects. Genome Res. 5 321–333. [PubMed]
  • Paterson, A. H., E. S. Lander, J. D. Hewitt, S. Peterson, S. E. Lincoln et al., 1988. Resolution of quantitative traits into Mendelian factors by using a complete linkage map of restriction fragment length polymorphisms. Nature 335 721–726. [PubMed]
  • Paterson, A. H., J. W. DeVerna, B. Lanini and S. D. Tanksley, 1990. Fine mapping of quantitative trait loci using selected overlapping recombinant chromosomes, in an interspecies cross of tomato. Genetics 124 735–742. [PMC free article] [PubMed]
  • Paterson, A. H., S. Damon, J. D. Hewitt, D. Zamir, H. D. Rabinowitch et al., 1991. Mendelian factors underlying quantitative traits in tomato: comparison across species, generations, and environments. Genetics 127 181–197. [PMC free article] [PubMed]
  • Paterson, A. H., Y. R. Lin, Z. Li, K. F. Schertz, J. F. Doebley et al., 1995. Convergent domestication of cereal crops by independent mutations at corresponding genetic loci. Science 269 1714–1718. [PubMed]
  • Paterson, A. H., J. E. Bowers, D. G. Peterson, J. C. Estill and B. A. Chapman, 2003. Structure and evolution of cereal genomes. Curr. Opin. Genet. Dev. 13 644–650. [PubMed]
  • Paterson, A. H., M. Freeling and T. Sasaki, 2005. Grains of knowledge: genomics of model cereals. Genome Res. 15 1643–1650. [PubMed]
  • Peleman, J. D., C. Wye, J. Zethof, A. P. Sørensen, H. Verbakel et al., 2005. Quantitative trait locus (QTL) isogenic recombinant analysis: a method for high-resolution mapping of QTL within a single population. Genetics 171 1341–1352. [PMC free article] [PubMed]
  • Redoña, E. D., and D. J. Mackill, 1996. Mapping quantitative trait loci for seedling vigor in rice using RFLPs. Theor. Appl. Genet. 92 395–402. [PubMed]
  • Redoña, E. D., and D. J. Mackill, 1998. Quantitative trait locus analysis for rice panicle and grain characteristics. Theor. Appl. Genet. 96 957–963.
  • Röder, M. S., X. Q. Huang and A. Börner, 2008. Fine mapping of the region on wheat chromosome 7D controlling grain weight. Funct. Integr. Genomics 8 79–86. [PubMed]
  • Konishi, S., T. Izawa, S. Y. Lin, K. Ebana, Y. Fukuta et al., 2006. An SNP caused loss of seed shattering during rice domestication. Science 312 1392–1396. [PubMed]
  • Sanguinetti, C. J., N. E. Dias and A. J. G. Simpson, 1994. Rapid silver staining and recover of PCR products separated on polyacrylamide gels. Biotechniques 17 915–919. [PubMed]
  • Song, X. J., W. Huang, M. Shi, M. Z. Zhu and H. X. Lin, 2007. A QTL for rice grain width and weight encodes a previously unknown RING-type E3 ubiquitin ligase. Nat. Genet. 39 623–630. [PubMed]
  • Takeuchi, Y., S. Y. Lin, T. Sasaki and M. Yano, 2003. Fine linkage mapping enables dissection of closely linked quantitative trait loci for seed dormancy and heading in rice. Theor. Appl. Genet. 107 1174–1180. [PubMed]
  • Takita, T., 1989. Breeding for grain shape in rice. Agri. Sci. 44 39–42.
  • Tan, Y. F., Y. Z. Xing, J. X. Li, S. B. Yu, C. G. Xu et al., 2000. Genetic bases of appearance quality of rice grains in Shanyou 63, an elite rice hybrid. Theor. Appl. Genet. 101 823–829.
  • Tan, Y. P., L. Z. Li, P. Li, L. X. Wang and Z. L. Hu, 2005. Quantitative trait loci for grain appearance traits of rice using a double haploid population. Mol. Plant Breed. 3 314–322.
  • Thomson, M. J., J. D. Edwards, E. M. Septiningsih, S. E. Harrington and S. R. McCouch, 2006. Substitution mapping of dth1.1, a flowering-time quantitative trait locus (QTL) associated with transgressive variation in rice, reveals multiple Sub-QTL. Genetics 172 2501–2514. [PMC free article] [PubMed]
  • Tsunematsu, H., A. Yoshimura, Y. Harushima, Y. Nagamura, N. Kurata et al., 1996. RFLP framework map using recombinant inbred lines in rice. Breed. Sci. 46 279–284.
  • Unnevehr, L. J., B. Duff and B. O. Juliano, 1992. Consumer Demand For Rice Grain Quality.International Rice Research Institute, Manila, The Philippines/International Development Research Centre, Ottawa, Canada.
  • Wan, X. Y., J. M. Wan, C. C. Su, C. M. Wang, W. B. Shen et al., 2004. a QTL detection for eating quality of cooked rice in a population of chromosome segment substitution lines. Theor. Appl. Genet. 110 71–79. [PubMed]
  • Wan, X. Y., J. F. Weng, H. Q. Zhai and J. M. Wan, 2004. b Fine mapping of pgwc-8 gene affecting percentage of grains with chalkiness in rice (Oryza sativa. L). Rice Genet. Newsl. 21 54–56.
  • Wan, X. Y., J. M. Wan, J. F. Weng, L. Jiang, J. C. Bi et al., 2005. Stability of QTLs for rice grain dimension and endosperm chalkiness characteristics across eight environments. Theor. Appl. Genet. 110 1334–1346. [PubMed]
  • Wan, X. Y., J. M. Wan, L. Jiang, J. K. Wang, H. Q. Zhai et al., 2006. QTL analysis for rice grain length and fine mapping of an identified QTL with stable and major effects. Theor. Appl. Genet. 112 1258–1270. [PubMed]
  • Wang, D. L., J. Zhu, Z. K. Li and A. H. Paterson, 1999. Mapping QTLs with epistatic effects and QTL by environment interactions by mixed linear model approaches. Theor. Appl. Genet. 99 1255–1264.
  • Wang, J. K., M. van Ginkel, R. Trethowan, G. Ye, I. DeLacy et al., 2004. Simulating the effects of dominance and epistasis on selection response in the CIMMYT wheat breeding program using QuCim. Crop Sci. 44 2006–2018.
  • Wang, J., S. C. Chapman, D. B. Bonnett, G. J. Rebetzke and J. Crouch, 2007. a Application of population genetic theory and simulation models to efficiently pyramid multiple genes via marker-assisted selection. Crop Sci. 47 580–588.
  • Wang, J. K., X. Y. Wan, H. H. Li, W. H. Pfeiffer, J. Crouch et al., 2007. b Application of identified QTL-marker associations in rice quality improvement through a design-breeding approach. Theor. Appl. Genet. 115 87–100. [PubMed]
  • Ware, D., and L. Stein, 2003. Comparison of genes among cereals. Curr. Opin. Plant Biol. 6 121–127. [PubMed]
  • Weng, J. F., X. Y. Wan, X. J. Wu, H. L. Wang, H. Q. Zhai et al., 2006. Stable expression of QTL for amylase content (AC) and protein content (PC) of milled rice (Oryza sativa L.) using a CSSL population. Acta Agron. Sin. 32 14–19.
  • Wu, J. Z., H. Mizuno, M. Hayashi-Tsugane, Y. Ito, Y. Chiden et al., 2003. Technical advance physical maps and recombination frequency of six rice chromosomes. Plant J. 36 720–730. [PubMed]
  • Wu, J. Z., T. Maehara, T. Shimokawa, S. Yamamoto, C. Harada et al., 2002. A comprehensive rice transcript map containing 6591 expressed sequence tag sites. Plant Cell 14 525–535. [PMC free article] [PubMed]
  • Xu, J. L., Q. Z. Xue, L. J. Luo and Z. K. Li, 2002. Genetic dissection of grain weight and its related traits in rice (Oryza sativa L.). Chinese J. Rice Sci. 16 6–10.
  • Xu, Z. J., W. F. Chen, D. R. Ma, Y. N. Lu, S. Q. Zhou et al., 2004. Correlations between rice grain shapes and main qualitative characteristics. Acta Agron. Sin. 30 894–900.
  • Yamamoto, T., Y. Kuboki, X. Y. Lin, T. Sasaki and M. Yano, 1998. Fine mapping of quantitative trait loci Hd-1, Hd-2 and Hd-3, controlling heading date of rice, as single Mendelian factors. Theor. Appl. Genet. 97 37–44.
  • Yamamoto, T., H. X. Lin, T. Sasaki and M. Yano, 2000. Identification of heading date quantitative trait loci Hd-6 and characterization of its epistatic interactions with Hd-2 in rice using advanced backcross progeny. Genetics 154 885–891. [PMC free article] [PubMed]
  • Yamanaka, N., S. Watanabe, K. Toda, M. Hayashi, H. Fuchigami et al., 2005. Fine mapping of the FT1 locus for soybean flowering time using a residual heterozygous line derived from a recombinant inbred line. Theor. Appl. Genet. 110 634–639. [PubMed]
  • Yano, M., and T. Sasaki, 1997. Genetic and molecular dissection of quantitative traits in rice. Plant Mol. Biol. 35 145–153. [PubMed]
  • Yeadon, P. J., F. J. Bowring and D. E. A. Catcheside, 2004. Alleles of the hotspot cog are codominant in effect on recombination in the his-3 region of neurospora. Genetics 167 1143–1153. [PMC free article] [PubMed]
  • Yoon, D. B., K. H. Kang, H. J. Kim, H. G. Ju, S. J. Kwon et al., 2006. Mapping quantitative trait loci for yield components and morphological traits in an advanced backcross population between Oryza grandiglumis and the O. sativa japonica cultivar Hwaseongbyeo. Theor. Appl. Genet. 112 1052–1062. [PubMed]
  • Zhang, Q. F., 2007. Strategies for developing Green Super Rice. Proc. Natl. Acad. Sci. USA 104 16402–16409. [PMC free article] [PubMed]
  • Zhang, Q. F., B. Z. Shen, X. K. Dai, M. H. Mei, M. A. S. Maroof et al., 1994. Using bulked extremes and recessive class to map genes for photoperiod-sensitive genic male sterility in rice. Proc. Natl. Acad. Sci. USA 91 8675–8679. [PMC free article] [PubMed]
  • Zhao, Z. G., C. M. Wang, L. Jiang, H. Ikehashi and J. M. Wan, 2006. Identification of a new hybrid sterility gene in rice (Oryza sativa L.). Euphytica 151 331–337.

Articles from Genetics are provided here courtesy of Genetics Society of America


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


  • EST
    Published EST sequences
  • MedGen
    Related information in MedGen
  • PubMed
    PubMed citations for these articles
  • Taxonomy
    Related taxonomy entry
  • Taxonomy Tree
    Taxonomy Tree

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...