RESUMEN

Accurate identification of plant diseases is important for ensuring the safety of agricultural production. Convolutional neural networks (CNNs) and visual transformers (VTs) can extract effective representations of images and have been widely used for the intelligent recognition of plant disease images. However, CNNs have excellent local perception with poor global perception, and VTs have excellent global perception with poor local perception. This makes it difficult to further improve the performance of both CNNs and VTs on plant disease recognition tasks. In this paper, we propose a local and global feature-aware dual-branch network, named LGNet, for the identification of plant diseases. More specifically, we first design a dual-branch structure based on CNNs and VTs to extract the local and global features. Then, an adaptive feature fusion (AFF) module is designed to fuse the local and global features, thus driving the model to dynamically perceive the weights of different features. Finally, we design a hierarchical mixed-scale unit-guided feature fusion (HMUFF) module to mine the key information in the features at different levels and fuse the differentiated information among them, thereby enhancing the model's multiscale perception capability. Subsequently, extensive experiments were conducted on the AI Challenger 2018 dataset and the self-collected corn disease (SCD) dataset. The experimental results demonstrate that our proposed LGNet achieves state-of-the-art recognition performance on both the AI Challenger 2018 dataset and the SCD dataset, with accuracies of 88.74% and 99.08%, respectively.

RESUMEN

River ecological health assessments are the premise for protecting river biodiversity and curbing river water environment deterioration. To explore the river ecosystem health status and its response mechanism to water environmental factors in the middle reaches of the Yarlung Zangbo River, phytoplankton samples were collected during July and October 2021, and water environmental factors were measured. The phytoplankton species were identified, and their cell abundance and biomass were calculated. The phytoplankton integrity index was constructed to evaluate the ecological health status of the river. The temporal and spatial distribution characteristics of water environmental factors and phytoplankton community, as well as the correlation between P-IBI value and community parameters and water environmental factors were analyzed. The results showed that the difference in water environmental factors was not evident in time but was significant in space. The average cell abundance and biomass of phytoplankton were in the order of wet season>dry season and main stream>tributaries. Diatoms dominated the community, and pH and WT were the major water environmental factors driving the spatial and temporal distributions of phytoplankton. The ecological health status of the middle reaches of the Yarlung Zangbo River was "healthy to sub-healthy." The river health status was better than that of the dry season, and that of the tributaries was better than that of the main stream. EC, TUR, WT, NO3--N, and NH4+-N were the major water environmental factors affecting the ecological health status of the river reach, which could be affected by the direction and rate of phytoplankton community succession. It is involved in and affects the process of material circulation and energy flow of the river ecosystem, thereby driving the ecological health of the middle reaches of the Yarlung Zangbo River.

Asunto(s)

RESUMEN

Eukaryotic microorganisms play an important role in the biogeochemical cycles of rivers. Dynamic hydrological processes in rivers are thought to influence the assembly processes of eukaryotic microbes, as well as affecting local geomorphology. These processes have not been extensively studied for eukaryotic river microbes in extreme environments on the Tibetan Plateau. This study used 18S rDNA gene amplification sequencing, a neutral community model, and a null model to analyze the spatial and temporal dynamics and assembly processes of eukaryotic microbial communities in the middle reaches of the Yarlung Zangbo River. We conducted analyses across wet and dry seasons, as well as varying altitudinal gradients. Our results showed that the diversity, structure, and taxonomic composition of eukaryotic microbial communities varied more with altitude than season, and the diversity of the communities first increased, then decreased, with increasing elevation. Distance-decay analysis showed that the correlation between eukaryotic microbial communities and environmental distance was stronger than the correlation between the microbial communities and geographical distance. Deterministic processes (homogeneous selection) dominated the construction of eukaryotic microbial communities, and water temperature, pH, and total phosphorus were the primary environmental factors that influenced the construction of eukaryotic microbial communities. These results expand our understanding of the characteristics of eukaryotic microbial communities in rivers on the Tibetan Plateau and provide clues to understanding the mechanisms that maintain eukaryotic microbial diversity in extreme environments.

Asunto(s)

RESUMEN

Soil salinity and alkalinity are important abiotic components that frequently have critical effects on crop growth, productivity and quality. Developing soybean cultivars with high salt tolerance is recognized as an efficient way to maintain sustainable soybean production in a salt stress environment. However, the genetic mechanism of the tolerance must first be elucidated. In this study, 257 soybean cultivars with 135 SSR markers were used to perform epistatic association mapping for salt tolerance. Tolerance was evaluated by assessing the main root length (RL), the fresh and dry weights of roots (FWR and DWR), the biomass of seedlings (BS) and the length of hypocotyls (LH) of healthy seedlings after treatments with control, 100 mM NaCl or 10 mM Na2CO3 solutions for approximately one week under greenhouse conditions. A total of 83 QTL-by-environment (QE) interactions for salt tolerance index were detected: 24 for LR, 12 for FWR, 11 for DWR, 15 for LH and 21 for BS, as well as one epistatic QTL for FWR. Furthermore, 86 QE interactions for alkaline tolerance index were found: 17 for LR, 16 for FWR, 17 for DWR, 18 for LH and 18 for BS. A total of 77 QE interactions for the original trait indicator were detected: 17 for LR, 14 for FWR, 4 for DWR, 21 for LH and 21 for BS, as well as 3 epistatic QTL for BS. Small-effect QTL were frequently observed. Several soybean genes with homology to Arabidopsis thaliana and soybean salt tolerance genes were found in close proximity to the above QTL. Using the novel alleles of the QTL detected above, some elite parental combinations were designed, although these QTL need to be further confirmed. The above results provide a valuable foundation for fine mapping, cloning and molecular breeding by design for soybean alkaline and salt tolerance.

Asunto(s)