Introduction
Yiyang Dahegu rice (YyDHG) is a specialty of Yiyang County, Shangrao City, Jiangxi Province, China, and belongs to the national agricultural product geographical indication (Rijin, Reference Rijin2009; Luo, Reference Luo2013). This rice variety, also known as ‘Magu’, is a unique japonica (Oryza sativa subsp. japonica) variety in the south and is specifically used for making rice cakes (Jongwen, Reference Jongwen1985; Xia et al., Reference Xia, Hua, Wang, Tang, Ting, Luo, Xiao, Wang, Chuanyuan, Liqing and Wang2021). The whole growth period of YyDHG is about 150 days, with a plant height of 130 cm. Its amylose content (AC) ranges from 14% to 16%, making its texture lie between glutinous rice and common japonica rice. Additionally, YyDHG has a tight plant type, moderate tillering ability, stout stems, straight flag leaves, long panicle necks, greenish-yellow leaf colour, short and round grains with brown long awns, yellowish-brown hulls and plenty of fuzz, showing good colour fading in the later stage (Kai et al., Reference Kai, Fang and Ren2017).
From a historical perspective, the cultivation of YyDHG can be traced back to the Tang Dynasty. After the autumn harvest at the end of each year, locals steam Dahegu rice (DHG) to make and process Yiyang rice cakes, a tradition that has been in existence for over 1,000 years. Yiyang rice cakes are renowned for their unique production techniques and high-quality DHG, making them one of the four famous rice varieties in Jiangxi Province (Chenxi, Reference Chenxi2021; Xia et al., Reference Xia, Hua, Wang, Tang, Ting, Luo, Xiao, Wang, Chuanyuan, Liqing and Wang2021) In summary, YyDHG is not only an important agricultural specialty of Yiyang County but also a crucial component of local cultural and economic development. Through continuous technological innovation and industrial upgrading, YyDHG and its derivative rice cake products are expected to maintain their market competitiveness.
Although YyDHG has significant value, research on it is still insufficient. YyDHG has remained popular for a thousand years, but little is known about the unique features that have contributed to its popularity. This research aims to address this knowledge gap by analysing and assessing the cultivar characteristics, genetic background and quality attributes of Yiyang Dahegu (YyDHG), along with the disparities between it and other Dahegu varieties (oDHG). Additionally, we aim to provide a scientific foundation for the conservation and advancement of this significant agricultural specialty product, facilitating the industrialisation and commercialisation of YyDHG and its derived products, while also contributing to the comprehension of the genetic diversity and evolutionary dynamics among rice varieties.
Materials and methods
Materials
We collected 89 samples of ‘DHG’ in Jiangxi Province, including 52 samples of YyDHG and 37 samples of oDHG (Supplementary Table S1), ensuring that all DHG resources in Jiangxi Province were collected.
Methods
Field management and trait measurement
From mid-December 2022 to the end of April 2023, the samples were planted in Sanya City, Hainan Province, with 8 columns and 2 rows for each sample, spaced at 16 cm × 20 cm. The field management methods were consistent with local standards. The observed traits included plant height, panicle type, leaf type, awn colour and panicle neck length, with measurement methods similar to those of (ZAHOOR, 2017). A total of 89 leaf samples were collected and stored at −20°C.
From mid-May to the end of October in both 2023 and 2024, the Gao’an Experimental Base in Jiangxi Province planted each sample in an 8-column by 12-row arrangement, with a spacing of 16 cm × 20 cm. The field management methods were consistent with local standards. Five plants were selected for phenotypic measurement, including traits such as heading date, leaf colour, leaf fuzz amount, panicle type, lodging resistance, stem angle, effective panicle number, glume hair, glume tip colour, awn colour, plant height, leaf length, leaf width, panicle neck length, SPAD value, flag leaf angle and glume colour. Some of the measurement methods were similar to those of Ma et al. (Reference Ma, Bing, Jianghong, Jiani, Cui, Leiyue, Huiying, Maomao and Longzhi2019) and Meng et al. (Reference Meng, Wang, Luo, Xu, Li, Diao, Gao, Hu and Zeng2021).
Variety identification by polyacrylamide gel electrophoresis and resequencing analysis
Leaves preserved in Sanya were used as experimental samples, and DNA was extracted using the Cetyltrimethylammonium Bromide/CTAB method (Murray and Thompson, Reference Murray and Thompson1980; Kuo et al., Reference Kuo, Henderson and Lambing2022). Twenty-four pairs of Simple Sequence Repeats/SSR molecular markers (Supplementary Table S2) were utilised to conduct polyacrylamide gel electrophoresis (PAGE) experiments (Bassam et al., Reference Bassam, Caetano-Anolles and Gresshoff1991; Hashemi et al., Reference Hashemi, Gharechahi, Nematzadeh, Shekari, Hosseini and Salekdeh2016).
In 2023, in Gao’an, 52 leaf samples were collected during the heading stage (Supplementary Table S1). These samples were preserved using liquid nitrogen and dry ice and sent to a company for next-generation sequencing. Resequencing: Illumina sequencing (150-bp paired-end) was performed on 52 accessions. Variant calling used GATK v4.2 with Q30 filtering (depth ≥ 10×). Post-sequencing analysis included the following: ① Variety identification was conducted using 48 pairs of SSR molecular markers from the national standard ‘NY/T 1433—2014’ (available at www.nssi.org.cn/nssi/front/85220410.html) (Supplementary Table S3). The specific steps involved sequencing the samples to obtain their nucleic acid sequences, aligning the primers to the nucleic acid sequences, identifying the sequence segment between the primers and then conducting pairwise comparisons of this sequence segment among different samples to identify differences. Finally, these differences were summarised into a difference matrix. If there are differences in the sequences between any two primers, these samples can be considered to belong to different varieties. ② Ten control samples (Supplementary Table S4) were introduced for population structure and kinship analysis, using the same method as described in Meng et al. (Reference Meng, Wang, Luo, Xu, Li, Diao, Gao, Hu and Zeng2021).
Analysis of quality traits
The quality traits assessed included appearance quality, milling quality, and cooking and eating quality. For appearance quality, experiments involved measuring grain length, grain width (GW) and 1,000-grain weight. For milling quality, experiments measured brown rice rate (BRR) and milled rice rate (MRR). Cooking and eating quality experiments focused primarily on measuring AC, gel consistency (GC), gelatinisation temperature (GT) and rapid viscosity analysis (RVA). RVA encompassed the following parameters: peak viscosity, hot paste viscosity, cool paste viscosity, breakdown value, setback value, consistence viscosity, pasting temperature and peak time. The measurement methods for these parameters were the same as those described in Zhou et al. (Reference Zhou, Zhang, Zhu, Peng, Chen, Fu, Ouyang, Bian, Hu and Sun2018).
Statistical analysis
Software Used: GraphPad Prism 9.0.0 (https://www.graphpad-prism.cn/). Analytical Methods: One-way Analysis of Variance and nonparametric tests were employed.
Results
Variety identification: The 89 samples consist of 52 varieties
Based on the combined results of PAGE experiments, field phenotypic observations and resequencing analysis, we found that the 89 samples consist of 52 varieties. Accurate variety identification is crucial for ensuring precise analysis of the genetic background, variety characteristics and quality traits of YyDHG. First, we analysed the electrophoresis bands (Supplementary Fig. S1) produced by the PAGE experiments for 24 pairs of SSR molecular markers (Supplementary Table S2). Then, based on the experimental results (Supplementary Tables S5 and S6), we initially inferred that the 89 experimental samples might contain 49 varieties. However, it is noteworthy that some SSR markers did not exhibit sufficient specificity or band clarity among the samples. To ensure the accuracy of variety identification, we decided to further evaluate the results in combination with field phenotypic observations.
Combining field observations, we found that there may be 52 varieties. During the field observations conducted in Sanya in 2023, we eliminated the sample ‘Yiyang Black Pearl’ based on the characteristic that DHG does not produce black rice. In the experimental fields of Gao’an, Jiangxi Province, in 2024 and 2025, through phenotypic comparisons (Table 1; Supplementary S7), we rejected our previous assessment of 49 varieties and confirmed that there may be 52 varieties among the 89 samples. Therefore, we need to revise the classification of some varieties. Specifically, DHG6-1 and DHG6-5 were separated, and DHG8-1, DHG8-2, DHG8-3 and DHG8-4 were each independently classified as separate varieties.
Phenotypic traits of representative varieties (2023)
Note: The values in the table are the means calculated from five repeated measurements.
We also adopted the national standard method: using the 48 pairs of SSR molecular markers specified in NY/T 1433—2014 (Supplementary Table S3) and combining the results of resequencing to review the variety identification results. Reassuringly, the review results were consistent with our field observations conducted in Gao’an in 2023 and 2024, confirming that the 89 samples indeed contained 52 different varieties (detailed data are provided in Supplementary Tables S8 and S9). This not only ensured the accuracy of variety identification but also laid a solid foundation for subsequent research.
Variety characteristics of YyDHG
Between 2023 and 2024, through detailed observations and records of YyDHG growing in the field (Supplementary Table S7), we were able to conduct an in-depth analysis of its main variety characteristics. The total growth period of YyDHG is approximately 150 days, with an average plant height of about 140 cm. It has a compact plant type, moderate tillering ability, sturdy stems and long panicle necks. In terms of leaves and grains, YyDHG has a greenish-yellow leaf colour, short and round grains with brown long awns and mustard-yellow husks covered with numerous hairs. These characteristics not only increase its identifiability but also reflect its unique variety characteristics.
Population analysis
We conducted a population analysis on the aforementioned 52 varieties, and the results showed that although there is abundant genetic diversity among DHG varieties, YyDHG did not exhibit significant subpopulation separation when compared with oDHG varieties. Further analysis of population structure and genetic relationships revealed two important findings: First, almost all DHG varieties tend to be japonica-type, consistent with previous studies (Xia et al., Reference Xia, Hua, Wang, Tang, Ting, Luo, Xiao, Wang, Chuanyuan, Liqing and Wang2021); secondly, all samples can be broadly classified into three major genetic groups.
To gain a more comprehensive understanding of the genetic evolutionary position of YyDHG within the rice species and its genetic relationships with other varieties, we specifically introduced 10 control samples, including 5 indica (O. sativa subsp. indica) and 5 japonica (O. sativa subsp. japonica) (details in Supplementary Table S4). These samples are widely distributed across China and have abundant closely related varieties and high representativeness. Through comprehensive analysis of these samples, we constructed a clear evolutionary tree, which showed two distinct indica and japonica subpopulations and confirmed that most DHG varieties belong to the japonica subpopulation. Notably, YyDHG did not show significant population separation from oDHG varieties within Jiangxi Province on the evolutionary tree. Based on comprehensive consideration of the phylogenetic tree of genetic relationships, cross-validation error analysis plots of population structure (Fig. 2a) and principal component analysis (PCA) results (Fig. 2c), we believe that these 52 varieties can be broadly classified into three groups (Fig. 2b).
Quality analysis
To facilitate comparative analysis, we divided these 52 rice varieties into three comparison groups based on whether they are YyDHG and their variety characteristics. The specific grouping is as follows: Group ① comprises 19 varieties of YyDHG; Group ② consists of 29 oDHG and Group ③ includes four varieties of DHG red rice. Subsequently, we measured the relevant traits related to their appearance quality, milling quality and cooking and eating quality. Finally, we conducted a thorough differential analysis of the collected data using GraphPad Prism 9.0.0 software.
Through careful observation of Fig. 3a–c, it is evident that there exist notable differences in appearance quality between YyDHG and oDHG, while significant disparities with DHG red rice are solely observed in GW. However, regarding milling quality (Fig. 3d,e) and cooking and eating quality (Fig. 3f–o), YyDHG does not demonstrate significant differences from oDHG, but significant differences are apparent across most measured values when compared to DHG red rice. These findings provide invaluable data support for our in-depth understanding of the quality characteristics of different varieties of DHG.
Discussion
The present study has made several key findings that significantly contribute to our understanding of the genetic background and quality characteristics of YyDHG.
Our genetic analysis revealed that, despite YyDHG being a cherished local agricultural specialty and cultural symbol, its genetic diversity does not significantly surpass that of other DHG varieties (oDHG). This observation is intriguing as it suggests that seed exchanges and cultivation practices between Yiyang and neighbouring regions have likely facilitated genetic homogenisation to some extent. This finding is novel as it challenges the common assumption that local specialties inherently possess unique genetic diversity due to their isolated cultivation histories.
We identified a cluster of four red rice varieties on the evolutionary tree, whose genetic characteristics aligned to the phenotypic traits of the indica subpopulation (detailed results in Fig. 1, Supplementary Table S7). This discovery provides further evidence of the complex genetic relationships among different rice types, transcending traditional classifications based on geographical origin or cultural significance.
Phylogenetic tree. Generated through the neighbour-joining method. The length corresponding to 0.05 in the figure serves as the scale for genetic distance.
The kinship heatmap analysis (Fig. 2d) indicated that while most DHG varieties exhibit close genetic distances on the phylogenetic tree, they do not display tight kinship among themselves. This result provides new insight into the genetic diversity and complexity within the DHG group. The results imply that despite sharing some genetic similarities, there are substantial genetic variations that could be leveraged in future breeding programmes to enhance traits such as yield, disease resistance and quality.
(a) Cross-validation error analysis diagram for population structure. CV stands for cross-validation error, k represents the assumed number of subgroups (or clusters). (b) Stacked chart of ancestral components. Different colours typically represent distinct ancestral populations or genetic components. k represents the number of potential ancestral components or populations that may exist. (c) PCA plot: 3D. PC1, PC2 and PC3 represent principal component 1, principal component 2 and principal component 3, respectively. The percentage figures indicate the proportion of the total variance explained by each principal component. (d) Heatmap of kinship. A colour gradient ranging from light blue to deep red indicates the change in relatedness from distant to close.
Box plot of DHG quality traits. ‘ns’ indicates no significant difference. *, ** and *** indicate significant differences at P < 0.05, P < 0.01 and P < 0.001, respectively. (a) Grain length/GL. (b) Grain width/GW. (c)Thousand-grain weight/TGW. (d)Brown rice rate/BRR. (e) Milled rice rate/MRR. (f) Amylose Content /AC. (g) Gel Consistency/GC . (h) Peak Viscosity. (i) Hot Paste Viscosity/HPV . (j) Breakdown Viscosity/BDV . (k) Setback Viscosity/SBV. (l) Cool Paste Viscosity/CPV. (m) Consistency Viscosity/CSV. (n) Pasting Temperature/PaT. (o) Peak Time/PeT.
There were no significant differences between YyDHG and oDHG for milling, cooking and eating quality. This suggests that the competitiveness of YyDHG in the market may not entirely depend on its unique quality characteristics, but rather more on its cultural value.
However, it is essential to acknowledge the limitations of our study that could be addressed in future research. The sample planting areas were confined to Gao’an and Sanya, which may introduce geographical biases into our conclusions. To obtain a more comprehensive understanding of YyDHG’s quality characteristics, the sample planting areas should be expanded to include Yiyang County and potentially other relevant regions. Moreover, traditional Yiyang rice cakes require a special production process involving ‘steaming three times and manually pounding 200 times’, which might contribute to their quality characteristics. However, this experiment did not address this aspect, as most Yiyang rice cakes on the market nowadays are machine-processed. However, this could be a topic of future research.
Looking ahead, future research should focus on harnessing the genetic diversity within the DHG group to develop improved rice varieties with enhanced traits. This could involve targeted breeding programmes that leverage the genetic differences identified in our study. In addition, more analytical approaches, such as transcriptomics and metabolomics, could be used to gain a deeper understanding of YyDHG’s genetic diversity and quality characteristics. Efforts should also be made to strengthen the protection of YyDHG varieties and the sustainable utilisation of germplasm resources. Increased research investment and technological innovation are crucial for promoting the sustainable development of the YyDHG industry. Furthermore, enhancing the promotion and marketing of YyDHG could boost its popularity and market competitiveness, thereby contributing to the prosperity of the local economy and culture.
Conclusion
Variety identification
Through PAGE experiments, field observations and resequencing analysis, a total of 89 samples were identified to include 52 different varieties, among which 19 were YyDHG, 29 were oDHG within Jiangxi Province and 4 were DHG Red rice.
Genetic evolution and kinship relationships
Population structure and kinship analysis have shown that almost all DHG varieties tend to belong to the japonica rice type. Moreover, there exists abundant genetic diversity among DHG varieties, yet there is no apparent subpopulation differentiation between YyDHG and oDHG within Jiangxi Province. Notably, the kinship heatmap reveals an interesting phenomenon: Although the vast majority of DHG varieties exhibit small genetic distances on the evolutionary tree, they do not display close kinship among themselves. In contrast, DHG red rice varieties clearly lean towards the indica rice community, showing significant differences from the japonica-type DHG varieties.
Quality characteristics
Regarding milling quality (BRR and MRR) and cooking and eating quality (AC, GC, GT, etc.), there were no significant differences between YyDHG and oDHG within Jiangxi Province, indicating that these varieties are relatively consistent in processing and eating quality.
However, significant differences were observed between YyDHG and DHG red rice in most quality indicators, particularly in GW, which may be related to the unique genetic background and growing environment of red rice varieties.
In conclusion, YyDHG, as a rice variety with local characteristics, exhibits unique values and advantages in terms of genetic diversity, kinship relationships, quality characteristics and market potential.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S1479262125100105.
Author contributions
L.Q.Y. and M.M.L. conceived and designed the experiments. B.X.M., J.X.H., J.L., H.T., W.J.L. and H.Y.Z. performed the experiments. B.X.M. analysed the data. W.H. and Y.Y. contributed samples. The manuscript was written by B.X.M., and L.Q.Y., M.M.L. and J.L. provided writing suggestions and revisions. All authors approved the final manuscript.
Funding statement
This work was supported by the National Natural Science Foundation of China (32360471) and the Jiangxi Provincial Subsidy Project for Modern Seed Industry Development.
Competing interests
The authors declare no competing interests.
Data availability
The resequencing data generated during the current study are not publicly available due to ‘Local crop resource protection policy’ but are available from the corresponding author on reasonable request.
Informed consent statement
Not applicable.
