Frontiers in Plant Science最新文献

Introduction: Plant disease segmentation in real-world agricultural environments poses significant technical challenges, including complex backgrounds, diverse lesion morphologies, and extreme class imbalance.

Methods: In this paper, we propose an integrated solution, STAR-Net, which combines a novel network architecture with a dynamic training strategy. The architecture features an innovative Heterogeneous Branch Attention Aggregation (HBAA) module to robustly represent multi-scale and multi-morphology features. The training strategy employs a Dynamic Phase-Weighted Loss (DPW-Loss) to navigate the complexities of imbalanced data.

Results: Our method achieves a state-of-the-art average mIoU of 93.36% on the NLB dataset. This result demonstrates its superior ability to precisely segment diseases with specific elongated morphologies. Furthermore, the model obtains a competitive average mIoU of 41.13% on the highly challenging PlantSeg dataset. This result validates its robustness in complex 'in-the-wild' scenarios.

Discussion: Our work presents a powerful, well validated, and synergistic solution for plant disease segmentation. It also paves the way for practical applications in precision agriculture.

As land salinization intensifies across the United States, there is growing interest in cultivating salt-tolerant crops such as olives (Olea europaea). With extra virgin olive oil becoming an increasingly popular and valuable commodity in the U.S. market, growing olive cultivars that can thrive under challenging conditions is critical. This study evaluated the physiological, antioxidative and biochemical responses of two relatively newly introduced olive cultivars, 'Oliana' and 'Lecciana', to salinity stress under controlled greenhouse conditions. Eight-month-old plants were subjected to three salinity treatments (0, 50, and 100 mM NaCl) in a completely randomized design (n = 9). Plant gas exchange parameters were measured at 0, 15, 30, and 45 days after treatment, while chlorophyll content, fluorescence (Fv/Fm), biomass, nutrient accumulation (in leaves, stems, and roots), antioxidant enzyme activity, osmolyte levels, and reactive oxygen species (ROS) concentrations were assessed at 15, 30, and 45 days. Salinity stress significantly reduced gas exchange parameters in 'Lecciana' cultivar as compared to 'Oliana', resulting in decreased chlorophylls, Fv/Fm, and nutrient content. In response, antioxidant enzymes and osmolyte accumulation (proline and glycine betaine) increased with 'Lecciana' showing the strongest response. Conversely, lipid peroxidation and hydrogen peroxide (H₂O₂) levels were highest in 'Oliana' under 100 mM NaCl, indicating greater oxidative stress. These findings suggest that 'Lecciana's high salinity tolerance is associated with enhanced antioxidant defense and compatible solutes. In addition, 'Lecciana' and 'Oliana' could be a promising cultivar for saline and drought-prone regions of the U.S., supporting the expanding production of high-quality extra virgin olive oil.

Far-red light (FR, 700-800 nm) can enhance photosynthesis by stimulating photosystem I (PSI). However, during chlorophyll fluorescence (CF) measurements using pulse-amplitude modulation (PAM) fluorometry, unusually high quantum yields of photosystem II (Φ _PSII) have been observed under high FR light intensities, raising concerns about measurement artifacts. To test this, we constructed light response curves for sweet basil (Ocimum basilicum L.) grown under light-emitting diode (LED) light (R:G:B = 44%:18%:38%) with varying photosynthetic photon flux densities (PPFD, 0-1,000 μmol m^-2 s^-1) and FR fractions (0, 0.26, 0.45, and 0.63). FR treatments consistently increased Φ _PSII, but when total photon flux density (TPFD, 400-800 nm) exceeded 1,000 μmol m^-2 s^-1, Φ _PSII rose abruptly. Nonfluorescent reference tests using white and black paper confirmed that FR induced spurious fluorescence signals, likely due to spectral overlap between FR photons and the PAM detection range (680-760 nm). Tilting the LED panel to reduce reflected FR eliminated the abrupt Φ _PSII peak but introduced unexpectedly increased Φ _PSII across treatments, likely due to probe-induced shading. These findings demonstrate that high-intensity FR can confound PAM-based CF measurements by producing spurious signals unrelated to plant physiology. Accurate and reliable assessment of photosynthetic performance under extended spectral lighting conditions requires careful management of lighting geometry and FR intensity.

Introduction: Complex traits arise from polygenic and interactive genomic architectures that are difficult to resolve using traditional genome-wide association study (GWAS) approaches. Machine learning (ML) provides complementary methods capable of capturing non-linear effects, improving signal detection, and enhancing predictive accuracy of marker trait associations (MTAs).

Methods: Using a publicly available winter wheat dataset (CIMMYT), we evaluated several widely used traditional GWAS tools, including GAPIT, GCTA, GEMMA, sommer, and TASSEL, with respect to computational efficiency, model performance, and the consistency of detected associations. In parallel, ML approaches, such as Elastic Net, Extreme Gradient Boosting (XGBoost), Random Forest, and the hybrid TSLRF model, were assessed based on feature importance metrics and functional annotation of selected markers.

Results: Despite a shared reliance on mixed linear models, the traditional GWAS tools exhibited differences in runtime and showed modest but meaningful variability in the number and overlap of MTAs. ML models recovered several associations detected by traditional methods and additionally identified novel markers, potentially reflecting non-linear or epistatic effects.

Discussion: Our findings demonstrate that ML can effectively complement traditional GWAS approaches for marker-trait identification in wheat. By extending beyond additive effects, ML broadens the scope of detectable genetic signals, providing a practical way to analyze complex traits and support informed marker-assisted breeding strategies.

Bowman-Birk protease inhibitors (BBIs) are multifunctional proteins with a double-headed structure, featuring two distinct inhibitory loops that target trypsin and chymotrypsin proteases. BBI regulates protease activity in plants and provides defense against pests and pathogens, but little is still known about their expression levels and their ability to interact with natural targets. Our results showed that BBI1 and BBI2 genes are the most highly expressed in Vigna seeds. Consequently, we produced two multiple sequence alignments including homologs from 42 Vigna taxa to explore variability and functionality. Phylogenetic relationships, signals of positive selection, and interaction energy levels with their natural targets were inferred. Overall, BBI2 exhibited the highest affinity for the assessed targets compared to BBI1. Amino acid substitutions have led to distinct protein variants across species, each displaying different interaction capacities with their respective targets. Additionally, the residue conferring inhibitory specificity for trypsin, located in the first domain, was found to be under positive selection in both genes. This suggests an ongoing evolutionary process aimed at optimizing affinity with proteases through continuous adaptation. Finally we emphasize that findings obtained can be used to drive the activity of plant breeders and more efficient cultivars can be selected. Given the growing availability of genomic information of wild and domesticated accessions, docking simulations offer a convenient and effective method to preliminarily assess new protein variants.

Ferruginous outcrops are ecologically formations that host high biodiversity and edaphic endemism. While canga outcrops in Carajás have been the focus of more extensive research, ferruginous outcrops in the Araguaia remain poorly studied, especially with respect to their functional ecology and conservation value. We evaluated the soils, floristic and functional compositions of plant communities on ferruginous outcrops in Carajás and the Araguaia, with the goals of comparing edaphic conditions, floristic compositions, and functional strategies between these disjunct regions and identifying patterns relevant for biodiversity conservation. A total of 129 plots were sampled spanning grassland (GS), shrubland (SB), and woodland (WD) formations. In all plots, soil samples were collected, and plant traits related to resource acquisition (SLA, leaf N, N:P), reproductive strategies (fruit dimensions), and interaction modes (dispersal and pollination syndromes) were measured. Herbaceous and woody communities (trees and treelets with dbh ≥= 3 cm) were analyzed separately. Functional similarity was assessed via community-weighted means and multivariate trait space analyses. Despite exhibiting moderate floristic similarity between regions and edaphic differences, both regions share acidic soils with low phosphorus (P) availability, a condition that imposes similar constraints on resource acquisition. Open formations (GS, SB) in both regions showed functional convergence, indicating similar environmental filters. In contrast, woody communities, especially those in WD, presented pronounced differences in trait composition, reflecting differences in local conditions and ecological history. This study highlights the complementary conservation value of ferruginous outcrops in Carajás and the Araguaia. The functional similarities in open formations suggest that these environments may exhibit ecological strategies associated with similar environmental conditions. Recognizing and protecting these unique environments is essential to ensure their long-term ecological resilience.

Introduction: Achillea fragrantissima is valued for its broad spectrum of secondary metabo-lites with notable therapeutic potential. This study examined the influence of an ethanolic extract of Commiphora myrrha gum-resin on in vitro shoot growth, callus formation, and the effect of compound composition in A. fragrantissima.

Material and methods: Shoot-tip and root-segment explants were cultured on Murashi-ge and Skoog (MS) medium supplemented with filter-sterilized C. myrrha ethanolic extract at 0.5, 1.0, and 2.0 g·L^-1, alongside control treatment. Plant responses, including growth parameters, callus induction percentage and biomass, photosynthetic pigments, and me-tabolite composition, were assessed after one month of in vitro culture. GC-MS analysis was performed on the C. myrrha extract and on methanolic extracts of A. fragrantissima plantlets and callus tissues obtained from different treatments.

Results: GC-MS analysis of the ethanolic extract of C. myrrha identified 66 phytochemi-cal compounds dominated by sesquiterpenes-particularly 2,5,8-trimethyl-1-nonen-3-yn-5-ol, curzerene, myrcenol, germa-cra-1(10),4,11(13)-trien-12-oic acid, and several oxygenated sesquiterpenes-indicating a rich pool of bioactive, growth-promoting constituents. Shoot-tip and root-segment ex-plants showed the strongest responses to C. myrrha extract at specific concentrations. The highest concentration tested (2.0 g·L^-1) produced the most shoots and the greatest accu-mulation of photosynthetic pigments, including total chlorophyll and carotenoids. In contrast, the lowest concentration (0.5 g·L^-1) promoted the longest shoots, the highest fresh weight, and the greatest number of leaves. Callus formation from shoot tips was also highest at 0.5 g•L^-1, while root-derived callus reached full induction at both 0.5 and 2.0 g·L^-1. Both shoot-tip plantlets and root-derived callus showed notable modifications in their bioactive constituents. Several bioactive constituents were notably elevated (e.g., de-sulphosinigrin, thymidine, 4H-pyran-4-one derivatives, and fatty acid derivatives).

Discussion: Distinct tissue-specific patterns emerged, shoot tips accumulated a wider range of aromatic and therapeutic compounds, whereas callus tissue was enriched in fat-ty acids. These findings suggest that C. myrrha extract functions as a natural biostimulant and elicitor, providing a sustainable approach for producing metabolite-rich A. fragrantis-sima material with potential pharmacological applications.

Background: Spodoptera frugiperda is a highly invasive pest that significantly threatens maize production globally. This work aims to report the physiological and biochemical impacts of different chemical treatments (DMSO, methyl oleate, DMF, xylene, uniconazole, azadirachtin, and chlorantraniliprole) on maize photosynthetic capacity and resistance response mechanisms from S. frugiperda.

Results: We found a dose-dependent effect on maize photosynthesis; lower concentrations of these chemicals promoted photosynthetic rate, whereas higher concentrations inhibited photosynthesis, especially in lower leaves. Mortality bioassays proved the dose-related response to the toxic potential of DMSO, DMF and xylene. However, the Y-tube bioassay revealed no remarkable changes in olfactory responses, thus indicating that the observed mortality was largely contributed by direct chemical toxicity rather than behavioral alterations. At the molecular level, cytochrome P450 genes (Sf CYP6AB12, Sf CYP6AE43, Sf CYP9A58 and Sf CYP9A59) were significantly overexpressed by chlorantraniliprole, and they were considered to be resistant genes against insecticides. Likewise, other compounds such as azadirachtin and uniconazole also selectively affected some P450 genes, providing additional evidence of the involvement of P450s in S. frugiperda metabolic resistance.

Conclusions: These results demonstrate the involvement of P450s in the development of insecticide resistance and suggest the importance of chemical dose on control of insect pests.

The main disease that affects chickpea production worldwide is Aschochyta blight (AB), caused by the fungus Aschochyta rabiei. The identification of cultivars with stacking resistance genes is crucial for controlling these diseases. This work aimed to evaluate the effect of stacking two resistance-related genes, chitinase and vst-1, on disease response in chickpea (Cicer arietinum L.). Gene stacking was achieved through conventional hybridization between three transgenic inbred lines: N292 and N346 (both carrying chitinase), and N52 (carrying vst-1). PCR confirmed the stable inheritance of both transgenes in F1 and F2 generations, although segregation ratios deviated from Mendelian expectations. Functional assays were conducted using protein extracts to test inhibition of fungal spore germination and mycelium formation, followed by detached-leaf and whole-plant infection assays. Protein extracts from stacked lines significantly reduced spore germination (up to 90% inhibition, P < 0.01) and suppressed mycelium development compared to controls. Detached-leaf assays revealed a reduced disease severity in stacked lines (mean DS = 74 vs. 89 in controls), while whole-plant assays confirmed lower severity scores (mean 4-6 vs. 8 in controls) despite no reduction in infection incidence. The hybrid N346 × N52 exhibited the strongest resistance phenotype across assays. These results demonstrate that stacking chitinase and vst-1 increases tolerance to A. rabiei in chickpea by reducing disease severity, providing a promising strategy for developing tolerant cultivars. This study is a successful tool for developing gene stacking technology in crops to contribute to improving the resistance of chickpea plants to Ascochyta disease.

Fruit recognition and ripeness detection are crucial steps in selective harvesting. To better address the difficulties of fruit recognition and ripeness detection techniques in complex facility environments, a novel lightweight tomato ripeness detection network model based on an improved YOLO v8s is proposed (called TRD-Net). Here, a tomato dataset including 3,330 images from real scenarios was constructed, and an accurate lightweight tomato ripeness detection model trained on the captured images was developed. The TRD-Net model achieves efficient detection of tomatoes affected by overlapping occlusions, lighting variations, and capture angles, offering swifter detection speeds and lower computational demands. Specifically, the feature extraction module of YOLO v8s was refactored by employing spatial and channel reconstruction convolution (SCRConv) and adding the SimAM attention mechanism. The CIoU loss function was replaced by the MPDIoU loss function. The performance of the novel TRD-Net was comprehensively investigated. The proposed TRD-Net achieved an mAP@0.5 of 0.9581 with an improvement of 4.32 percentage points, and the model size decreased from 22.5 M to 17.6 M with an inference time of 8.7 ms per image. The number of model parameters and floating-point operations per second (FLOPs) decreased by 19.69% and 22.03%, respectively. Compared with state-of-the-art models, the proposed TRD-Net is notably promising for real-time tomato recognition and maturity detection. The study contributes to the establishment of a machine vision sensing system for a selective harvesting robot in a complex gardening environment.