Foods with a low glycemic index and rich antioxidant components are increasingly valued for their health benefits. This study quantified sucrose, glucose, and fructose levels and characterized the metabolic profiles of brown sweet and non-sweet rice. Sweet rice exhibited the higher sucrose and glucose content compared to non-sweet rice. Metabolomic analysis identified 1,347 annotated metabolites, including 490, 503, and 267 differentially accumulated metabolites (DAMs) between sweet and non-sweet rice. Furthermore, comparative analyses among sweet rice revealed 305, 355, and 409 DAMs, with the most significantly enriched pathways being flavonoid biosynthesis, tryptophan metabolism, and flavone and flavonol biosynthesis. Notably, sucrose and glucose levels were elevated in sweet rice, accompanied by substantial increases in bioactive compounds and nutrients, including essential amino acids, flavonoids and phenolic acids. Elucidating the metabolites and associated metabolic pathways of brown sweet and non-sweet rice provides valuable insights for molecular breeding and the nutritional enhancement of rice varieties.

Single-cell transcriptome sequencing (scRNA-seq) can reveal the roles of diverse cells in an organism, but accurately classifying cell subpopulations and their marker genes remains a challenge. Here, we present PhytoCell, an ensemble learning framework that combines feature selection engineering with machine learning to uncover cell markers and annotate cell subpopulations. We evaluated our approach on 120,000 cells from corollas of the dicotyledonous plant species coyote tobacco (Nicotiana attenuata) and eight tissues from the monocotyledonous plant species rice (Oryza sativa). Comprehensive evaluation across species and tissues demonstrated that PhytoCell effectively eliminates redundant information, identifies key cell markers, improves clustering performance, and accurately classifies cell subpopulations. Importantly, PhytoCell did not rely on prior biological knowledge for selecting cell markers, preserving the biological landscape of the original data. For broader accessibility, we developed a user-friendly web interface that provides convenient tools for users to access cell marker resources and perform predictions for cell type. PhytoCell is freely accessible at https://cgris.net/phyto. PhytoCell is scalable to different sizes of single-cell datasets, representing a valuable resource for precise identification in cell research.

The semi-dwarf cereal varieties bred during the Green Revolution revolutionized global agriculture under optimal growing conditions, but their performance in stressful environments-, particularly under soil salinity, has remained an unresolved paradox. Here, we show that Green Revolution varieties (GRVs) of rice and wheat exhibit significantly enhanced salt tolerance compared with their pre-Green Revolution cultivated counterparts (non-GRVs), mediated by stress-induced accumulation of DELLA proteins. Through integrated metabolomic and transcriptomic analyses, we demonstrate that DELLAs maintain "growth-stress" balance by rewiring sugar-amino acid metabolic networks. At the molecular level, DELLAs antagonize INDETERMINATE SPIKELET1 (IDS1), a growth-promoting transcription factor that impairs salt tolerance through biomolecular condensation. Structural and functional analyses demonstrate that DELLAs physically dissolve IDS1 condensates, thereby reprogramming transcriptional networks. Remarkably, expression of a dominant-negative OsIDS1 variant (OsIDS1EARm), which attenuates condensation and transcriptional repression, confers both semi-dwarf architecture and enhanced salt tolerance in non-GRVs, outperforming conventional Green Revolution alleles by producing a 35% yield gain (∼170 kg ha⁻¹) in saline fields. Collectively, our work resolves the mechanistic basis of stress adaptation in semi-dwarf crops and establishes a novel paradigm for the development of stress-resilient crops through targeted manipulation of transcriptional condensates.

Accurate estimation of rice yield traits from intact panicles is important for modern crop breeding. Here, non-threshing rice panicle phenotyping refers to quantitative trait measurement without detaching or threshing individual grains before imaging. Red–green–blue (RGB) imaging provides detailed morphology but struggles to distinguish filled and unfilled grains with similar appearance, whereas thermal infrared (TIR) imaging provides filling-related thermal contrast but limited structural detail. To address these limitations, we developed Panicle-Net, an RGB–TIR pipeline for estimating seed setting rate (SSR) and thousand-grain weight (TGW) from intact rice panicles. Its perception module, ZR-YOLO, integrates RGB and TIR features through a Bidirectional Iterative Feature Fusion Module (BIFM) to detect and classify individual grains. SSR was calculated from frame-level filled/unfilled counts averaged across a 10-frame cooling sequence, whereas TGW was predicted from sequence-level morphological features of detected filled grains. On the independent test set, RGB-only and TIR-only models achieved mean average precision at an intersection-over-union threshold of 0.5 (mAP@0.5) values of 82.3% and 80.1%, respectively. ZR-YOLO reached 93.7%, improvements of 11.4 and 13.6 percentage points. The complete pipeline achieved an SSR mean absolute error (MAE) of 1.77 percentage points with an R² of 0.96, and a TGW MAE of 0.73 g with an R² of 0.89. These results demonstrate the effectiveness of RGB–TIR fusion for non-threshing intact-panicle phenotyping.

Salt stress is one of the major abiotic factors limiting rice yield, with the tillering stage—an essential growth phase that strongly influences rice productivity—being particularly sensitive to salinity. Thus, identifying salt-tolerant rice varieties is of great importance for ensuring stable rice production. In this study, we systematically evaluated the salt tolerance of 372 rice landraces at the tillering stage through dynamic phenotypic monitoring, using the average salt injury score (ASIS) as an indicator at two (T2W) and four weeks (T4W) after salt treatment. A genome-wide association study (GWAS) identified 39 loci significantly associated with salt tolerance. Among these, two high-confidence candidate genes, OsST8.1 and OsST8.2, both members of the BTB-MATH protein family, were implicated in salt tolerance during the tillering stage. Haplotype analysis revealed significant differences (p < 0.05) in salt tolerance among germplasm carrying different haplotypes, with accessions harboring the superior haplotype exhibiting enhanced tolerance. Consistently, qRT-PCR analysis showed significantly lower or higher expression levels (p < 0.05) of OsST8.1 or OsST8.2 in accessions with the superior haplotype following salt treatment, suggesting that they may regulate rice responses to salinity stress. Through the validation of a knockout transgenic experiment, OsST8.2 was identified as the causal gene for salt tolerance in rice at the tillering stage. Collectively, this study provides valuable genetic resources and a theoretical foundation for elucidating the genetic basis of salt tolerance and for breeding new salt-tolerant rice varieties.

Introduction:Key rice panicle traits (grain number, panicle length, grain dimensions, maturity) determine yield and quality, and high-precision/high-throughput measurement is critical for rice breeding. Traditional methods are. Methods:A dataset of 5300 rice panicle images (loose/normal/dense types; milk/dough/full maturity/over-ripe stages) was constructed, with 3290 for training, 940 for validation, and 470 for testing. A deep learning pipeline integrating. Results:The panicle length extraction achieved R²=0.9583, RMSE=5.69 mm. Grain counting R² values were 0.9799 (loose), 0.9551 (normal), 0.9278 (dense). Grain length R²=0.8823, grain width MAPE=6.64%. OPG-YOLOv8. Discussion:This study provides a comprehensive, automated tool for rice panicle phenotyping, addressing occlusion challenges and bridging the gap between advanced models and breeding applications.

Cold sensitivity in rice leads to significant yield losses. Identifying cold-tolerant germplasm and uncovering cold tolerance genes are essential for developing resilient rice varieties. Kam Sweet Rice (KSR) is notable for strong cold tolerance, which is an invaluable genetic resource for identifying such genes. In this study, we phenotyped cold tolerance across various growth stages using 104 KSR accessions and 268 other rice landraces. Genome-wide association studies (GWAS) identified 89 loci significantly associated with cold tolerance: 57 loci at the germination stage, 9 at the bud stage, and 28 at the seedling stage, with 61 loci (69%) being newly discovered. Through association and selection sweep analyses, we identified two high-confidence candidate genes, OsCTD2 and OsLTPL159, associated with cold tolerance at both germination and seedling stages. Haplotype analysis revealed significant differences in cold tolerance grade and survival rate among various haplotypes of these genes, with superior haplotypes predominantly present in KSR. RNA-seq and qRT-PCR results showed that the superior haplotypes of OsCTD2 and OsLTPL159 exhibited significantly higher expression in cold-tolerant accessions under cold stress, whereas no significant differences were observed in cold-sensitive accessions with the inferior haplotypes. These results indicated that OsCTD2 and OsLTPL159 are involved in cold stress response in rice. Additionally, we identified two other promising candidate genes and their superior haplotypes for cold tolerance: OsGRS7 at the germination stage and OsBSR6 at the bud stage. Our findings provide a solid foundation for cloning cold tolerance genes and offer insights for designing molecular breeding strategies.

Rice is highly sensitive to low temperatures, making cold stress a significant factor limiting its growth, especially during the bud bursting stage. To address this, an RIL population derived from a cross between cold-tolerant and cold-sensitive rice varieties was used to identify nine QTLs linked to cold tolerance under temperatures of 4 ℃, 5 °C, and 6 ℃ using a high-density genetic map. One candidate gene, LOC_Os07g44410, was identified through gene function annotation, haplotype analysis, and qRT-PCR, with two main haplotypes (Hap1 and Hap2) showing distinct phenotypic differences. qRT-PCR analysis showed that the expression level of LOC_Os07g44410 in cold tolerant lines carrying Hap1 was significantly higher than that in cold sensitive lines carrying Hap2. Hap1, associated with greater cold tolerance, was predominant in japonica rice, while Hap2 related to cold sensitive was majority in indica rice. This study offers valuable genetic resources for further research on cold tolerance mechanisms and breeding applications at the bud bursting stage in rice.

Submergence tolerance QTLs for rice germination were identified via a genome-wide association study, and a new causal gene, LOC_Os06g17260, was identified. Submergence stress is a major obstacle limiting the application of direct seeding in rice cultivation. Rapid bud and root growth helps plants establish a stronger growth base and improve their submergence tolerance. Therefore, mining genes for bud length (BL) and root length (RL) helps in the development of varieties that are adaptable to submergence and improve seedling emergence and yield of direct-seeded rice. In this study, a genome-wide association study of BL and RL was performed on a diverse rice collection consisting of 300 accessions. We identified a total of 37 QTLs, 13 of which had phenotypic contributions > 10

The global demand for nutrient-rich functional rice is rapidly increasing. However, the metabolic mechanisms underlying the high-nutrient functionality of special rice remain inadequately understood. In this study, we developed 26 novel special rice accessions. Using a widely targeted metabolomics approach, we identified and classified 1347 metabolites. The metabolite profiles of these accessions exhibited substantial differentiation, with 204–629 differentially accumulated metabolites identified. Notably, pigmented giant embryo glutinous (PGEG) rice integrated a diverse array of bioactive compounds and nutrients from pigmented rice and giant embryo glutinous rice. It was particularly enriched in γ-aminobutyric acid (GABA), flavonoids, phenolic acids, alkaloids, and other physiologically active substances. Key metabolic pathways (including the biosynthesis of flavone, flavonol, and flavonoid) were significantly enriched in differential metabolites between white/glutinous and PGEG rice. These findings provide a comprehensive understanding of the metabolite profiles of special rice varieties and offer valuable insights for breeding future high-nutrient functional rice.

Single-cell RNA sequencing (scRNA-seq) technology enables a deep understanding of cellular differentiation during plant development and reveals heterogeneity among the cells of a given tissue. However, the computational characterization of such cellular heterogeneity is complicated by the high dimensionality, sparsity, and biological noise inherent to the raw data. Here, we introduce PhytoCluster, an unsupervised deep learning algorithm, to cluster scRNA-seq data by extracting latent features. We benchmarked PhytoCluster against four simulated datasets and five real scRNA-seq datasets with varying protocols and data quality levels. A comprehensive evaluation indicated that PhytoCluster outperforms other methods in clustering accuracy, noise removal, and signal retention. Additionally, we evaluated the performance of the latent features extracted by PhytoCluster across four machine learning models. The computational results highlight the ability of PhytoCluster to extract meaningful information from plant scRNA-seq data, with machine learning models achieving accuracy comparable to that of raw features. We believe that PhytoCluster will be a valuable tool for disentangling complex cellular heterogeneity based on scRNA-seq data.

Providing reliable and sufficient food resources for a growing global population is a significant societal challenge, in particular achieving this while keeping nitrogen pollution within safe environmental limits. Therefore, it is crucial to identify nitrogen-use efficient (NUE) accessions and identify candidate genes associated with NUE for sustainable agricultural development. Here, we present analysis on the genetic diversity of 518 accessions of Chinese germplasm and performed a genome-wide association study (GWAS) on 16 traits associated with NUE and yield. We identified a total of 89 significant loci, including 47 associated with NUE and 42 with yield, of which 56 (63 %) were newly discovered. Through association and indica-japonica genetic differentiation analysis, we identified a high-confidence candidate gene - OsNPT4 - that encoding a protein from the POT family. This gene was associated with nitrogen grain production efficiency (NGPE), nitrogen harvest index (NHI), and tillering number (TN). RNA-seq results indicated that OsNPT4 may play a crucial role in effectuating the response of rice plants to nitrogen treatment. Further haplotype analysis revealed significant differences among the various haplotypes of this gene concerning NGPE, NHI, and TN, with accessions carrying Hap1 demonstrating strong NUE and increased yields. RT-qPCR results showed that OsNPT4 expression significantly increased in Hap1-carrying accessions in both leaves and roots upon treatment, while no significant differences were observed in Hap2-carrying accessions. This further confirmed OsNPT4 as a key candidate gene associated with varying NUE. Taken together, our results provide a theoretical foundation for cloning NUE genes and facilitate the design of molecular breeding strategies.

Improving grain quality is second only to enhancing grain yield in breeding hybrid rice. Yet, rice grain quality, especially milling and appearance quality, is facing increasing threats from global warming due to climate change, leading to a relatively slow progress in high-quality rice breeding. Identifying additional grain quality genes is an effective way to combat against the threats on rice grain quality. In the present study, we used the germplasm from 3,000 Rice Genomes Project for genome-wide association study, and identified GL7/GW7/SLG7 as a major QTL besides GS3 and GW5 for grain shape. Among nine haplotypes of GL7 (H1-9), H1-H4, which harbored an 11-bp deletion, were designated as the functional GL7 allele and were primarily present in Geng/Japonica (GJ) rice. We developed KASP markers for GL7 major haplotypes (H1 and H2), and established a breeding system assisted by the markers to effectively improve Xian/Indica (XI) hybrids for grain quality. Yuehesimiao (YHSM) and Yixiang 1 A (YX1A) are widely applicated XI restorer line and sterile line in three-line hybrids. The improved hybrid YX1AGL7/YHSMGL7 exhibited longer grain, higher ratio of grain length-to-width and larger rate of head rice, but lower chalkiness rate and degree than that of any other hybrids and both parents. Furthermore, the improved hybrid with GL7 had statistically same yield of grains with all other hybrids, indicating no penalty of grain yield while improving grain quality. The GL7 haplotype along with its marker KASP-S2 and breeding strategy resulted from this study could be valuable sources for developing XI hybrids with high quality and high yield of grains in rice.

Rice grain quality improvement is closely linked to economic development, local dietary habits, and traditional culture. Despite significant advances in breeding over the past decade, enhancing grain quality remains a major challenge. Here, we evaluated 315 newly sequenced rice varieties (163 japonica and 152 indica) to assess milling quality (e.g., head rice yield, HRY), cooking quality (e.g., amylose content, AC), and nutritional quality (e.g., protein content, PC). Clear regional differences were observed: Yunnan and Jiangsu contributed more high-quality japonica varieties, whereas Guangdong and Guizhou produced more high-quality indica varieties. A genome-wide association study identified several QTLs and putative candidate genes associated with grain quality traits. A composite haplotype (Hap1) of the Wx gene, defined by both coding and intronic polymorphisms, was associated with low AC and high HRY. Additional putative candidate genes were suggested, including LOC_Os04g40760 and LOC_Os04g40780 (HRY), LOC_Os06g04080 (AC), and LOC_Os06g20020 (PC), with expression differences observed across growth stages. Superior haplotypes, such as Hap1 of LOC_Os04g40760 and LOC_Os04g40780, are rare in wild rice but have increased in frequency during domestication, indicating strong selection. Collectively, these findings highlight regional disparities in the prioritisation of grain quality traits and provide preliminary genomic resources and hypotheses for future functional studies and marker-assisted breeding.

Mesocotyl length is a key trait affecting seedling emergence and establishment in dry direct-seeded rice, with longer mesocotyls promoting rapid and uniform emergence, thereby forming larger effective populations. Therefore, mining genes associated with mesocotyl length will facilitate the development of rice varieties suitable for dry direct seeding. In this study, 300 rice germplasm resources with a wide range of sources were selected as experimental materials. Phenotypic traits such as mesocotyl length and seedling emergence rate were systematically determined in each variety by setting different mulch depth treatments. Genome-wide association analysis (GWAS) was used to locate QTL controlling mesocotyl length and predict candidate genes. The results showed that mesocotyl length increased significantly with greater soil cover depth, while excessively deep sowing treatments inhibited seedling emergence. The GWAS analysis identified four QTLs associated with mesocotyl length and two QTLs associated with seedling emergence, with phenotypic contributions of 6.96-8.48%. Among them, the mesocotyl length-related QTL qML3 located at 28.03-28.43 Mb on chromosome 3 was detected at both sowing depths. Gene annotation analysis identified nine candidate genes related to plant hormones and transcription factors for qML3. Further investigation revealed three genes (LOC_Os03g49250, LOC_Os03g49400, and LOC_Os03g49510) exhibiting distinct haplotypes with significant differences in mesocotyl length, suggesting they may be causal genes for qML3. The results provide new clues to elucidate the molecular mechanism of rice mesocotyl development and lay an important foundation for subsequent gene function verification and molecular breeding. In the future, the functions of these candidate genes will be verified by transgenic and other methods, and molecular markers will be developed for genetic improvement of drought-tolerant rice varieties.

Modern cultivated rice plays a pivotal role in global food security. China accounts for nearly 30% of the world's rice production and has developed numerous cultivated varieties over the past decades that are well adapted to diverse growing regions. However, the genomic bases underlying the phenotypes of these modern cultivars remain poorly characterized, limiting the exploitation of this vast resource for breeding specialized, regionally adapted cultivars. In this study, we constructed a comprehensive genetic variation map of modern rice using resequencing datasets from 6044 representative cultivars from five major rice-growing regions in China. Our genomic and phenotypic analyses of this diversity panel revealed regional preferences for specific genomic backgrounds and traits, such as heading date, biotic/abiotic stress resistance, and grain shape, which are crucial for adaptation to local conditions and consumer preferences. We identified 3131 quantitative trait loci associated with 53 phenotypes across 212 datasets under various environmental conditions through genome-wide association studies. Notably, we cloned and functionally verified a novel gene related to grain length, OsGL3.6. By integrating multiple datasets, we developed RiceAtlas, a versatile multi-scale toolkit for rice breeding design. We successfully utilized the RiceAtlas breeding design function to rapidly improve the grain shape of the Suigeng4 cultivar. These valuable resources enhance our understanding of the adaptability and breeding requirements of modern rice and can facilitate advances in future rice-breeding initiatives.

Weedy rice, a wild relative of cultivated rice, is highly stress-resistant and proliferates in paddy fields. In this study, 353 weedy rice accessions were analyzed to identify salt-tolerance genes using population evolution analysis, phenotypic screening, genome-wide association studies (GWAS), transcriptome analysis, haplotype characterization, gene knockout experiments, and Na+ and K+ ion flux assays. Population structure analysis classified the accessions into six distinct groups. Three salt-tolerant accessions-HW131, HW136, and HW119-were identified based on leaf rolling degree (LRD), leaf withering degree (LWD), chlorophyll content (ChlC), and nitrogen content (NC) traits. GWAS and transcriptome data pinpointed LOC_Os06g39270 and LOC_Os06g11860 as candidate salt-tolerance genes. Haplotype analysis and qPCR confirmed two major haplotypes: AHap2 and BHap1. A 2-bp deletion (TC) at position 818 bp in LOC_Os06g11860 was associated with severe salt sensitivity (phenotypic grade 7), whereas the wild-type exhibited strong tolerance (grade1). Knockout mutants exhibited significantly increased Na+ and K+ flux across mesophyll cell membranes compared to wild-type plants, validating LOC_Os06g11860 (OsERFH1) as a crucial salt-tolerance gene. This study provides novel genetic insights into salt-stress adaptation in weedy rice, paving the way for breeding enhanced salt-tolerant varieties.

Grain shattering is one of the critical traits in rice, influencing harvesting methods and yield. Seeds from varieties that shatter easily often begin to drop before reaching full maturity. Varieties with difficult-to-thresh or non-threshing characteristics are susceptible to having their branches broken and mixed with straw during mechanical harvesting, leading to relatively severe losses. Therefore, developing rice varieties with a moderate degree of shattering, suitable for production applications, is the primary focus in rice breeding. In this study, the F-2:3 generation population from a cross between the easy-shattering weedy rice variety "H21" and the non-shattering japonica rice variety "Longdao 18" was utilized to investigate shattering resistance through phenotypic identification, Bulk Segregant Analysis (BSA), gene editing, and cytological detection. Gene Os01g0935000, a ZOS1-23 - C2H2 zinc finger protein involved in the regulation of shattering, was preliminarily localized and named OsZIPH1. Mutant phenotype identification showed that the shattering rate of the mutant is 41.98 %, while that of the wild type is 84.33 %, the decrease is 50.2 % compared to the wild type H21. Cytological analysis revealed that mutant spikelets had incomplete abscission layer structures between the lemma and rachis branches, whereas wild-type plants exhibited clearly defined abscission layer structures. The expression level of this gene in the wild type is 2.63 times that in the knockout mutant. Haplotype analysis has revealed that this gene comprises five haplotypes. HAP5 is unique to Oryza rufipogon, while HAP1 is found in nearly all japonica rice varieties. HAP2 is predominantly present in 70 % of indica weedy rice, and HAP3 is primarily associated with 59.37 % of indica improved rice. HAP4 is found in SH and BHA weedy rice, indicating that the shattering gene Os01g0935000 plays a significant role in the domestication process of rice. These findings enhance our understanding of the biological mechanisms underlying rice shattering and are crucial for developing rice varieties with desirable shattering characteristics.

The tolerance of rice to drought and saline stress is crucial for maintaining yields and promoting widespread cultivation. From an ethyl methanesulfonate (EMS)-mutagenized mutant library, we identified a mutant that is susceptible to osmotic stress, named Osmotic Stress Sensitivity 1 (Oss1). Using MutMap sequencing, we characterized the role of a choline transporter-related family gene, CTR4 (Choline Transporter-Related 4), in rice’s tolerance to drought and salt stress. CTR4 plays a critical role in regulating membrane lipid synthesis. In knockout mutants, the total membrane lipid content, especially unsaturated fatty acids, was significantly reduced. Compared with the wild type, knockout mutants exhibited decreased membrane lipid stability under drought and salt stress, faster water loss, higher relative electrolyte leakage, and lower levels of proline and soluble sugars, leading to impaired tolerance to drought and salt stress. In contrast, the overexpression of CTR4 enhanced seedling tolerance to drought and saline stress. The overexpression lines displayed lower malondialdehyde levels, reduced relative electrolyte leakage, and slower rates of leaf water loss under stress conditions, thereby improving seedling survival rates during stress. Moreover, lipid synthesis gene expression was down-regulated in CTR4 mutants, potentially exacerbating membrane permeability defects and further compromising stress resistance. These findings suggest that CTR4 mediates choline transport and influences cell membrane formation, thereby enhancing rice defenses against drought and salt stress by maintaining lipid homeostasis.