Plant Methods最新文献

{"title":"Non-destructive monitoring of root biomass in hydroponically grown leafy vegetables: comparison between machine learning-based RGB and hyperspectral imaging.","authors":"Ziyi Jin, Daisuke Yasutake, Tadashige Iwao, Yuki Sago, Gaku Yokoyama, Shigehiro Kubota, Tomoyoshi Hirota","doi":"10.1186/s13007-026-01515-8","DOIUrl":"10.1186/s13007-026-01515-8","url":null,"abstract":"<p><strong>Background: </strong>Root biomass serves as a critical indicator of plant eco-physiological status and crop productivity, yet its non-destructive monitoring remains challenging because of its underground location. The use of transparent nutrient film technique (NFT) systems enables direct observation of entire root systems, rendering image-based phenotyping feasible. In this study, we investigated and compared the performance of RGB and hyperspectral imaging for predicting root dry weight in hydroponically grown spinach (Spinacia oleracea L.).</p><p><strong>Results: </strong>Using 430 root segments divided from 60 plants, three models were developed: (1) an area-based regression based on root coverage, (2) a convolutional neural network (CNN) using RGB images, and (3) a partial least squares regression (PLSR) model using hyperspectral data (450-950 nm). The area-based regression exhibited limited accuracy (R² = 0.446) because of saturation at high root coverage. The CNN model improved predictive performance (R² = 0.739) but tended to overestimate sparse roots as a result of resolution constraints. The PLSR model achieved the highest accuracy (R² = 0.822, RMSE = 0.019 g/segment), with significantly lower error than RGB-based approaches (P < 0.01). Variable importance in projection analysis indicated that PLSR effectively exploited spectral signatures at 450 nm (background contrast) and 750 nm (tissue scattering), thereby maintaining stable accuracy across the full biomass range. When validated using 104 independent plants, the PLSR model achieved high predictive accuracy. Furthermore, as a proof of concept, this model successfully visualized the spatiotemporal dynamics of root biomass accumulation over 50 days, with only a 7.70% relative error at harvest.</p><p><strong>Conclusions: </strong>To our knowledge, this study is among the first to demonstrate the non-destructive monitoring of biomass distribution within entire root systems under production conditions. Hyperspectral imaging combined with PLSR outperforms RGB-based approaches by capturing spectral signatures that reflect internal tissue properties of roots, thereby overcoming limitations caused by morphological occlusion. This approach provides a robust tool for precision agriculture and high-throughput phenotyping, enabling continuous assessment of root growth through simple modifications to the existing hydroponic systems.</p>","PeriodicalId":20100,"journal":{"name":"Plant Methods","volume":" ","pages":""},"PeriodicalIF":5.2,"publicationDate":"2026-03-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13088466/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147434693","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An efficient clear-native PAGE-based workflow for cryo-electron microscopy sample preparation of large protein complexes.","authors":"Zitong Yang, Shinsa Kameo, Soichiro Seki, Genji Kurisu, Ryouichi Tanaka, Akihiro Kawamoto, Atsushi Takabayashi","doi":"10.1186/s13007-026-01513-w","DOIUrl":"10.1186/s13007-026-01513-w","url":null,"abstract":"<p><strong>Background: </strong>Cryo-electron microscopy (cryo-EM) has revolutionized protein research by enabling high-resolution structural analysis. However, preparing ultra-large protein complexes (e.g., > 700 kDa) for cryo-EM remains challenging, as it requires preserving both structural integrity and the native state. Conventional isolation methods, such as sucrose density gradient centrifugation, require large sample volumes and provide limited separation resolution. In contrast, native PAGE offers higher resolution; however, no established method exists for extracting protein complexes from gels followed by further purification to achieve high purity. Consequently, no standardized native PAGE-based protocol for cryo-EM sample preparation avoids multiple purification steps. Hence, we aimed to develop a rapid and efficient cryo-EM protein sample preparation method using electroelution with an optimized buffer system that preserves complex integrity to recover target protein complexes after sodium deoxycholate (DOC)-based clear-native PAGE (CN-PAGE).</p><p><strong>Results: </strong>We developed an agarose-acrylamide composite gel, which is simpler to prepare and mechanically more robust than conventional linear-gradient acrylamide gels commonly used for CN-PAGE, facilitating precise band excision for efficient electroelution. Cryo-EM structural analysis of the photosystem I-light-harvesting complex I (PSI-LHCI) supercomplex from Arabidopsis thaliana achieved high resolution (2.18 Å) after electroelution from this gel, requiring only buffer exchange by ultrafiltration to remove DOC before grid preparation, without additional chromatographic purification. This finding suggests that DOC may be the main inhibitor of successful grid preparation.</p><p><strong>Conclusion: </strong>Our results demonstrate the potential of this method for isolating large protein complexes from small sample volumes for cryo-EM structural analysis. This approach may expand the scope of cryo-EM targets to include some challenging systems previously hindered by purification difficulties.</p>","PeriodicalId":20100,"journal":{"name":"Plant Methods","volume":" ","pages":""},"PeriodicalIF":5.2,"publicationDate":"2026-03-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13088791/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147434640","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A high-throughput fluorescence-based microplate reader assay to quantify total flavonol levels in plant tissues.","authors":"Hana Daryanavard, Joëlle K Muhlemann","doi":"10.1186/s13007-026-01511-y","DOIUrl":"10.1186/s13007-026-01511-y","url":null,"abstract":"<p><strong>Background: </strong>Flavonols are plant specialized metabolites that regulate plant growth, development, and stress responses. Due to their antioxidant activity, they also confer nutritional benefits to human health. Quantification of flavonols in plant tissues typically relies on chromatographic methods such as HPLC or LC-MS, or on microscopy-based approaches using diphenylboric acid 2-aminoethyl ester (DPBA) staining to visualize flavonols in plant tissues. These methods are time-consuming and resource intensive. Here, we present a rapid, high-throughput, fluorescence-based microplate reader assay for flavonol quantification, in which the flavonol-specific dye DPBA is added to plant extracts to form fluorescent complexes.</p><p><strong>Results: </strong>The assay was optimized for extraction efficiency and validated for sensitivity, accuracy, and reproducibility. It also shows consistency with HPLC measurements. We demonstrate its utility by quantifying flavonol levels in different tomato tissues, across different cultivars, and even between plant species. Our assay showed that reproductive tissue in tomato plants has higher flavonol levels than vegetative tissue. Also, we found variation in flavonol levels in tomato fruit skin across different laboratory and commercial cultivars, suggesting that our approach shows promise for use in genome-wide association studies to identify genetic factors underlying variation in flavonol levels. Lastly, we measured flavonol levels in dry leaves of different plants used for brewed beverages.</p><p><strong>Conclusion: </strong>In conclusion, the assay represents a simple, robust, and scalable flavonol screening tool for studies in plant metabolism, environmental physiology, breeding, and metabolic engineering.</p>","PeriodicalId":20100,"journal":{"name":"Plant Methods","volume":" ","pages":""},"PeriodicalIF":5.2,"publicationDate":"2026-03-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13081301/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147369965","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"FE-SEM visualization of cortical microtubules in plant cells using freeze-fracture techniques.","authors":"Tomohiro Hatano, Satoshi Nakaba, Hikaru Nagata, Iwao Doi, Yoshiki Horikawa, Satoshi Kimura, Ryo Funada","doi":"10.1186/s13007-026-01510-z","DOIUrl":"10.1186/s13007-026-01510-z","url":null,"abstract":"<p><strong>Background: </strong>Cortical microtubules (CMTs), one of the components of cytoskeleton, control the orientation and localization of newly deposited cellulose microfibrils in cell walls, and thereby determine the shape, size, and structure of plant cells. Imaging of CMTs in plant tissues is generally performed using fluorescently labeled specimens under an optical fluorescence or confocal laser scanning microscope. However, visualizing individual CMTs by optical microscopy is challenging because the observable range is limited to superficial tissue layers accessible to light penetration. In contrast, transmission electron microscopy offers high-resolution visualization of CMTs in plant cells but is restricted to slightly oblique ultrathin sections with an approximate thickness of 70-100 nm. Given that field emission scanning electron microscopy (FE-SEM) provides both high spatial resolution and a broad field of view, the establishment of FE-SEM-based techniques for CMTs visualization is highly desirable.</p><p><strong>Results: </strong>Herein, we introduce a technique for visualizing CMTs within untreated plant tissues by combining two cryo-fracture techniques, freeze-knife fracture and freeze-tensile fracture, with FE-SEM. We successfully observed the arrangement of CMTs in several plant specimens, including young branches of ginkgo (Ginkgo biloba), calli from the leaves of hybrid poplar (Populus sieboldii × P. grandidentata), and root tips of the adzuki bean (Vigna angularis). CMTs were visualized on the protoplasmic fracture face (PF) using both cryo-FE-SEM and conventional room-temperature FE-SEM. While cryo-FE-SEM was employed to maintain native structures, room-temperature FE-SEM proved to be a viable alternative when specialized cryo-equipment was unavailable.</p><p><strong>Conclusions: </strong>The combination of freeze-fracture techniques with FE-SEM enables the visualization of CMT arrangement in plant tissues at a high resolution and across a broad area without the need for staining or extraction of cellular components. This technique is applicable to various plant tissues and allows for detailed observation of CMTs within these tissues, providing valuable insights into the role of microtubules in the division and differentiation of plant cells.</p>","PeriodicalId":20100,"journal":{"name":"Plant Methods","volume":" ","pages":""},"PeriodicalIF":5.2,"publicationDate":"2026-03-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13067565/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147348641","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Comparative evaluation of physical methods for extracting leaf trichomes.","authors":"Satinderpal Kaur, Devi Balakrishnan, Rupesh Kariyat","doi":"10.1186/s13007-026-01512-x","DOIUrl":"10.1186/s13007-026-01512-x","url":null,"abstract":"<p><p>Trichomes are hairlike protuberances in plants that deter herbivores using both physical and chemical means, with pre- and post-ingestive effects. In addition, trichomes also serve as a protective layer, reducing transpiration loss, thereby improving abiotic stress tolerance. A common method for assessing the role of trichomes in defense is to extract and infuse trichomes into the artificial diet, and then examine the feeding, growth, and developmental parameters of herbivores. Although methods such as brushing, using dry ice, and shaving with a razor, have been commonly used, there is no consensus on the best and most efficient method for trichome extraction. In the current study, to optimize the ideal method to extract trichomes, we tested three common methods with some modifications, such as flash freezing and brushing, shaving using a scalpel, and vortexing followed by freezing. Here, we used representative species from both eudicots and monocots with significant trichome diversity. For eudicots, we selected three cultivated plant species: tomato, sunflower, and Mexican squash, and two wild plant species: wild sunflower and wild squash, and for monocots, we used the model plant, rice. The efficiency of each method was examined using a combination of light and scanning electron microscopy. The results suggest that freezing followed by brushing is the most efficient and ideal method for extracting trichomes and can be recommended for eudicots. However, none of the evaluated methods were effective for the monocot rice.</p>","PeriodicalId":20100,"journal":{"name":"Plant Methods","volume":" ","pages":""},"PeriodicalIF":5.2,"publicationDate":"2026-02-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13041295/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147309068","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A simple and high-throughput single pollen genotyping pipeline to study meiosis in apple (Malus x domestica).","authors":"Han Palmers, Lucas Philips, Nico De Storme","doi":"10.1186/s13007-026-01502-z","DOIUrl":"10.1186/s13007-026-01502-z","url":null,"abstract":"<p><strong>Background: </strong>Pollen are the primary carrier of male genetic information during sexual reproduction. In Angiosperms, pollen are produced in the anthers following the process of meiosis, during which ploidy halving and genetic recombination occurs, mixing the parental genomes and creating allelic variation. Therefore, the genetic configuration of pollen is of great interest for the study of meiotic recombination and cell division as well as genetic variation in general, having applications in plant genetics, breeding, and evolution. Single pollen genotyping recently emerged as a powerful tool for the genetic characterization of the male germline, allowing a more detailed analysis of genetic diversity and reproductive behavior down to the level of the individual gamete. However, current methods for single pollen genotyping rely on manual sorting and use of expensive whole genome amplification technology.</p><p><strong>Results: </strong>We developed a simple and straightforward pipeline for single pollen genotyping in apple, an economically important fruit crop. In brief, this method includes filter bursting to efficiently release pollen nuclei followed by isolation of single nuclei via fluorescence-activated cell sorting. Subsequently, single nuclei are directly PCR genotyped, resulting in up to 85% amplification success without the use of whole genome amplification. Using this optimized pipeline, we developed a multiplex genotyping toolbox for apple which can be used to study meiotic recombination, meiotic restitution, and cross-over interference.</p><p><strong>Conclusions: </strong>Overall, this new single pollen genotyping procedure for apple enables high-throughput genotyping of large pollen populations in a single cell manner, facilitating reproductive research in apple as a non-model crop species with potential for translation to other fruit species.</p>","PeriodicalId":20100,"journal":{"name":"Plant Methods","volume":" ","pages":""},"PeriodicalIF":5.2,"publicationDate":"2026-02-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13037024/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147276815","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"tRNA-based polycistronic CRISPR/Cas9 system boosts efficiency of multi-gene deletion in the moss Physcomitrella.","authors":"Elena Kozgunova","doi":"10.1186/s13007-026-01509-6","DOIUrl":"10.1186/s13007-026-01509-6","url":null,"abstract":"<p><p>CRISPR/Cas9-based genome editing in the model bryophyte Physcomitrium patens (commonly known as Physcomitrella) is widely used for gene knockout via small insertions or deletions (indels). In this study, we developed an efficient dual-gRNA system capable of producing large, targeted deletions across multiple genes, enabling straightforward detection by gel electrophoresis and simultaneous multi-gene knockout. We first compared the efficiency of polycistronic tRNA-gRNA arrays to conventional gRNA constructs expressed under individual promoters, using the checkpoint protein gene MAD2 as a target. We found that a polycistronic construct doubled the frequency of large gene deletions compared to a conventional design. We then demonstrated that simultaneous deletion of two or four genes, targeting the katanin and TPX2 gene families, respectively, can be achieved in a single transformation event. The polycistronic system also increased deletion frequencies in the multiplex context, with up to 42% efficiency for individual genes and successful recovery of quadruple mutants. As a drawback, we confirmed that deletion efficiency varied substantially among individual gRNA pairs, indicating that gRNA design remains a critical factor in multiplex editing. This study establishes a fast and efficient framework for simultaneous removal of multiple genes in Physcomitrella, providing a practical alternative to homologous recombination-based methods for functional and applied studies.</p>","PeriodicalId":20100,"journal":{"name":"Plant Methods","volume":" ","pages":""},"PeriodicalIF":5.2,"publicationDate":"2026-02-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13032600/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146776708","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Real-time identification and quantification of apple scab on fruit in preharvest and postharvest conditions using YOLO11: a deep learning approach.","authors":"Fernanda Leiva, Séverine Gabioud Rebeaud, Danilo Christen","doi":"10.1186/s13007-026-01508-7","DOIUrl":"10.1186/s13007-026-01508-7","url":null,"abstract":"<p><strong>Background: </strong>Apple scab (AS), caused by the fungal pathogen Venturia inaequalis, is a major disease of apple that manifests as lesions on leaves and fruits. The disease compromises fruit quality and yield, leading to substantial economic losses. Traditional AS assessment relies on visual scoring, which is labor-intensive, subjective, and poorly reproducible. This study proposes a deep learning-based framework to overcome these limitations and to enable an accurate, scalable AS phenotyping approach.</p><p><strong>Results: </strong>Deep learning techniques were employed for the object detection and segmentation of AS symptoms in apple fruits. A two-stage fine-tuning process was applied to color images collected under orchard and laboratory conditions using the YOLO foundation model (YOLO11). The model was first trained to detect healthy apple fruits (Model 1) and subsequently refined to segment AS lesions (Model 2) using high-resolution imagery (864 × 864 pixels). Model 1 (Fruit Detection) achieved 0.98 precision, 0.95 recall, and 0.94 mAP50. Model 2 (Lesion Segmentation) achieved 0.64 precision, 0.75 recall, and 0.75 mAP50. The framework supports real-time processing of images and video. Despite challenges such as variable lighting and symptom heterogeneity, the use of high-resolution training data improved the segmentation accuracy (mAP50-95) of fine-scale lesions by over 50% compared to the previous YOLO architecture.</p><p><strong>Conclusion: </strong>These results demonstrate that the proposed deep learning-based approach provides a reliable pipeline for automated AS phenotyping. By improving precision and efficiency in both controlled and field environments, the model enhances apple grading assessments and accelerates breeding efforts to identify AS-resistant genotypes. Furthermore, this work establishes a solid foundation for broader applications in real-time plant disease monitoring and future integration of additional apple diseases.</p>","PeriodicalId":20100,"journal":{"name":"Plant Methods","volume":" ","pages":""},"PeriodicalIF":5.2,"publicationDate":"2026-02-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13032425/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147271907","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"SCFM-DETR: an enhanced transformer-based method for automated maize disease detection in field environments.","authors":"Sasa Tian, Zhiqing Tao, Ke Li, Yuan Rao, Xianhong Xie, Yuan Yuan, Jun Zhu","doi":"10.1186/s13007-026-01507-8","DOIUrl":"10.1186/s13007-026-01507-8","url":null,"abstract":"<p><p>Maize is susceptible to various diseases throughout its growth cycle, which can significantly reduce yields. The accurate identification of maize diseases with similar symptomatic manifestations is particularly challenging under field conditions due to heterogeneous lighting and variable weather conditions. This paper proposes a novel detection model named SCFM-DETR, which is based on an improved Real-Time DEtection TRansformer (RT-DETR) to achieve robust identification of maize diseases in complex environments. SimAM-StarNet is employed as the backbone for feature extraction in this model, reducing the number of parameters and improving multiscale feature fusion, thereby diminishing the impact of background noise. Furthermore, the original RepC3 module is replaced with a newly designed CGLU-FasterBlock-MANet (CFM) module, which enhances adaptive feature fusion for finer discriminative capability. The experimental results demonstrate that the SCFM-DETR model achieves an average precision of 96.7% and a recall of 95.8% on a maize disease dataset, exceeding the corresponding metrics of the baseline RT-DETR-R18 model by 3.1% and 6.0%. Additionally, the model reduces the number of parameters and computational load by 47% and 49%, respectively, making it highly suitable for deployment in computationally limited agricultural settings. This work offers a high-accuracy, lightweight framework that facilitates intelligent crop disease monitoring and supports the advancement of smart agriculture.</p>","PeriodicalId":20100,"journal":{"name":"Plant Methods","volume":" ","pages":""},"PeriodicalIF":5.2,"publicationDate":"2026-02-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13011694/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146202419","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Paft-wpest: wolfberry pests fine-grained classification method based on generative self-supervised learning.","authors":"Jianping Liu, Yue Zhang, Jianhua Zhang, Jian Wang, Guomin Zhou, Wei Sun, Libo Liu, Haiyu Ren, Xi Chen, Pan Liu","doi":"10.1186/s13007-026-01506-9","DOIUrl":"10.1186/s13007-026-01506-9","url":null,"abstract":"<p><p>Fine-grained pest recognition is a key component of intelligent pest monitoring and precise control, and it is important for ensuring agricultural production safety. This paper proposes a generative self-supervised learning-based pest recognition model, termed PAFT-WPest, to address challenges in fine-grained pest recognition, including small inter-class differences, large intra-class variations, complex background interference, and limited annotated data. The model employs partial-convolution spatial attention to focus on pest regions while suppressing redundant background information. Channel semantic selection and frequency-domain modeling are introduced to enhance the model's ability to perceive subtle detail differences. In addition, the model captures dependency relationships among different parts of the pest body to improve the modeling of global structure and semantic information. Furthermore, two fine-grained wolfberry pest datasets that distinguish pest growth stages and damage locations are constructed, and a continual pre-training strategy is adopted to enhance cross-scenario adaptability. Experimental results show that PAFT-WPest achieves accuracies of 76.83%, 91.53%, 98.70%, 79.27%, and 97.34% on the public pest datasets IP102, Butterfly-200, WPIT9K, Rice Pest, and Jute Pest, respectively, and accuracies of 97.82% and 94.69% on the self-built wolfberry pest datasets WP45 and WP11. These results indicate that the proposed model can improve fine-grained pest recognition performance under complex backgrounds, providing a feasible approach for agricultural pest monitoring and classification.</p>","PeriodicalId":20100,"journal":{"name":"Plant Methods","volume":" ","pages":""},"PeriodicalIF":5.2,"publicationDate":"2026-02-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12977878/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146143235","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}