Plant Methods最新文献

Generating accurate and visually realistic 3D models of plants from single-view images is crucial yet remains challenging due to plants' intricate geometry and frequent occlusions. This capability matters because it supplements current plant datasets and enables non-destructive, high-throughput phenotyping for crop breeding and precision agriculture. More broadly, 3D reconstruction is particularly important because plant morphology is inherently three-dimensional, while 2D representations miss occluded leaves, branching geometry, and volumetric traits. However, plants present unique challenges compared to common rigid objects, and most current generative methods have not been systematically tested in this domain, leaving a gap in understanding their reliability for realistic plant reconstruction. This study systematically evaluates six advanced generative techniques-Hunyuan3D 2.0, Trellis (Structured 3D Latents), One2345++, InstantMesh, Direct3D and Unique3D-using the existing PlantDreamer dataset. Specifically, this research reconstructs mesh models from images of Bean plants and quantitatively assesses each method's performance against ground-truth models using Chamfer Distance, Normal Consistency, F-Score, PSNR, LPIPS, and CLIP Score. The paper also presents qualitative results of Kale and Mint plants. The results indicate that Hunyuan3D 2.0 achieves superior performance overall, suggesting its effectiveness in capturing complex plant structures. This work provides valuable insights into strengths and limitations of contemporary 3D generative approaches, guiding future improvements in realistic plant digitisation.

Microscopic imaging provides essential visual evidence for pathogen monitoring, but its shallow depth of field and the three-dimensional height variation of spores lead to pronounced defocus blur and structural degradation. Restoring such images is therefore crucial for reliable spore identification and downstream analysis. However, microscopic defocus is a spatially varying process that severely suppresses high-frequency structures, causing natural-image deblurring models to generalize poorly. In addition, optical constraints of microscopy make realistic sharp-blur pairs difficult to obtain, further limiting learning-based restoration. To address these challenges, we propose MicroDeblurNet, a single-image deblurring network specifically designed for microscopic defocus restoration. The model incorporates a convolutional block attention module to enhance spatial selectivity toward key pathogen structures, and employs depthwise over-parameterized convolutions to capture locally varying blur patterns more effectively, enabling spatially consistent and structurally coherent restoration. Furthermore, a spatial-frequency consistency loss is proposed to strengthen high-frequency detail recovery while maintaining color fidelity and morphological integrity. To support high-fidelity supervision, we propose a paired-data construction strategy based on Laplacian-pyramid fusion and construct a clear-blur microscopic dataset for cucumber downy mildew. The restored outputs of MicroDeblurNet are further applied to sporangia detection and semantic segmentation to evaluate their impact on high-level visual tasks. Finally, we build an integrated microscopic analysis platform that delivers standardized high-quality data and automated pathogen-structure recognition and analysis to support disease assessment and management. Experimental results demonstrate that MicroDeblurNet achieves an optimal balance across pixel-level, structure-level, and perception-level metrics, reaching a PSNR of 42.48 dB and SSIM of 0.9839, outperforming advanced state-of-the-art methods. In downstream tasks, MicroDeblurNet delivers higher detection recall and segmentation accuracy in challenging scenarios involving sporangia adhesion and background impurities, demonstrating its ability to enhance target discernibility, preserve structural completeness, and improve robustness under complex microscopic conditions.

Background: Reliable classification of rose cultivars is complicated by their long and complex breeding history, frequent hybridization, and the coexistence of traditional horticultural categories with genetically heterogeneous groups. While molecular marker sets such as SSRs have been applied to assess genetic relationships, studies across cultivated roses and species are rare. Using 18 SNP markers on 1,345 accessions in combination with machine learning now offers an opportunity to systematically evaluate how well horticultural classes align with underlying genomic structure and to provide robust tools for the management of large germplasm collections.

Results: Using a panel of SNP markers across 1,345 rose accessions from the Europa Rosarium Sangerhausen, multiple unsupervised (hierarchical, spectral, k-means, DBSCAN, HDBSCAN) and supervised (SVM, decision tree, naive Bayes, XGBoost) machine learning approaches were applied to identify genetic clusters and predict horticultural classifications. Across clustering methods, certain groups consistently emerged as genetically distinct, such as the alba and damask roses, which clustered together with low internal diversity, reflecting their shared historic origin. In contrast, tea, bengal, lutea, and remontant hybrids were repeatedly grouped together and predicted with high classification accuracies (up to 100%) but displayed high within-group diversity, which is consistent with complex breeding backgrounds. Miniature, kordesii, and rubiginosa hybrids also tended to cluster together, despite their differing horticultural labels. Overall, the labels obtained from unsupervised clustering were consistently confirmed by supervised models, which achieved balanced accuracies of up to 100%, highlighting the robustness of the observed groupings.

Conclusions: Our results demonstrate that machine learning applied to SNP marker data can robustly resolve genetic relationships among rose cultivars and provide novel insights into the alignment of horticultural classifications with genomic structure. The high predictive accuracies obtained suggest that marker-based classification can serve as a reliable complementary tool for genebank management, cultivar identification, and reassessment of traditional rose categories.

Background: The root-knot nematode Meloidogyne incognita, is a highly destructive parasite that manipulates host plant processes through effector proteins, affecting agriculture globally. Despite advances in genomic and transcriptomic studies, the regulatory mechanisms controlling effector gene expression, especially at the chromatin level, are still poorly understood. Gene regulation studies in plant-parasitic nematodes (PPN) face several challenges, including the absence of transformation systems and technical barriers in chromatin preparation, particularly for transcription factors (TFs) expressed in secretory gland cells. Conventional methods like Chromatin Immunoprecipitation (ChIP) are limited in PPN due to low chromatin yields, the impermeability of nematode cuticles, and difficulties in producing antibodies for low-abundance TFs. These issues call for alternative approaches, such as dCas9-based CAPTURE (CRISPR Affinity Purification in siTU of Regulatory Elements) that allows studying chromatin interactions by using a catalytically inactive dCas9 protein to target specific genomic loci without relying on antibodies.

Results: This study presents an optimized in vitro dCas9-based CAPTURE for second stage juvenile (J2) M. incognita that addresses key challenges in chromatin extraction and stability. The protocol focuses on the promoter region of the 6F06 effector gene, a critical gene for parasitism. Several optimizations were made, including improvements in nematode disruption, chromatin extraction, and protein-DNA complex stability. This method successfully isolated chromatin-protein complexes and identified four putative chromatin-associated proteins, including BANF1, linked to chromatin remodelling complexes like SWI/SNF.

Conclusion: The optimized in vitro dCas9-based CAPTURE protocol offers a new tool for investigating chromatin dynamics and regulatory proteins in non-transformable nematodes. This method expands the scope of effector gene regulation research and provides new insights into M. incognita parasitism. Future research will aim to validate these regulatory proteins and extend the method to other effector loci, potentially guiding the development of novel nematode control strategies.

Background: Single-cell transcriptomics is a powerful approach to resolve cellular heterogeneity, yet its application in plants is constrained by challenges in tissue preparation, nuclei isolation, and transcriptome quality. Optimized experimental and computational workflows are essential to achieve robust results in plant systems.

Results: We systematically benchmarked bulk and single-cell transcriptomic workflows in maize and established an integrated, optimized framework. First, we developed an improved bulk RNA-seq protocol, providing higher consistency and serving as a reference for single-cell datasets. Second, we compared three input types, protoplasts, fresh nuclei, and frozen nuclei, across tissues, demonstrating overall comparability of their transcriptomic profiles and offering guidance for studies with limited material. Third, by leveraging bulk RNA-seq as a reference, these complementary data provide additional biological context that helps to interpret and validate findings derived from single-cell transcriptomic analyses. A combination of these strategies resulted in high transcriptome integrity and clear clustering resolution in the final dataset, supporting robust identification of plant cell types. While all experimental data are derived from maize, the principles and strategies described here provide practical guidance and inspiration for single-cell studies in other plant species.

Conclusions: Our study establishes optimized experimental and computational workflows for plant single-cell transcriptomics. By validating input comparability and addressing the limitations of nuclear data, we provide methodological guidance that extends beyond maize and supports future single-cell investigations across diverse plant species.

Background: In Shandong Province of China, where annual precipitation is below 800 mm, tea plants face persistent drought stress exacerbated by global warming. Breeding drought-tolerant tea cultivars is one of the effective ways to cope with this challenge. However, traditional breeding approaches are still limited by prolonged cycles, low efficiency, and subjective evaluation. To overcome these limitations, the development of rapid and objective germplasm evaluation methods has become critical‌.

Results: In this study, hyperspectral images of leaves from 12 widely cultivated 'Lucha series' tea cultivars in Shandong Province during different drought periods were collected, and the drought-related physiological indicators were measured simultaneously. Then, a tea drought tolerance index (TDTI) with enhanced accuracy was established by integrating the rate of change of indicators with temporal weights and indicator weights. Subsequently, we developed a novel lightweight Transformer-based hybrid integrated architecture to establish prediction models for the physiological indicators and TDTI. The Transformer-based models synergistically combined a Transformer encoder with XGBoost and LightGBM within a lightweight framework that leverages ensemble learning, data augmentation, and regularization to ensure robustness on limited datasets. Finally, we compared the performance of Transformer-based models against traditional machine learning models.​ The optimal models for MRP, MDA, Pro, SS, ChlT and TDTI were identified as 1D-CARS-TF, 2D-UVE-SVM, 2D-UVE-BRR, 2D-CARS-SVM, 1D-UVE-TF-CNN, and 2D-UVE-TF, respectively, achieving determination coefficient (R²) of 0.8992, 0.8307, 0.8929, 0.8373, 0.7894, and 0.7614, on an independent test set.​ The results demonstrated that the lightweight Transformer-based models equipped with multi-head self-attention mechanism exhibited outstanding capabilities in processing indicators requiring multi-band correlation mining. ​​Simultaneously, feature selection algorithms and overfitting-mitigation optimization strategies played a critical role in enhancing both the accuracy and stability of the Transformer-based models.​.

Conclusions: This study established a robust technical foundation for rapid, accurate, and non-destructive comprehensive evaluation of drought tolerance for tea plant germplasm resources. However, it should be noted that they were based on a specific set of greenhouse-cultivated samples, and further validation under field conditions with expanded germplasm resources would strengthen generalizability. Anyway, the demonstrated potential of the Transformer-based model in our study advances phenomics of tea plants toward greater intelligence and efficiency.