江西农业大学

The precise identification of safe navigable areas constitutes a foundational requirement for autonomous driving in complex unstructured environments. Existing studies have predominantly concentrated on extracting local or single-scale features, neglecting the substantial complementarity between multi-scale representations in complex natural environments, which constrains semantic understanding. To address this issue, a Semantic-Aware Terrain Segmentation Network (SATSNet) is proposed for navigable area recognition with guidance-entropy, which incorporates global sparse contextual information as prior guidance and progressively explores the latent complementarity among multi-scale features from both semantic hierarchy and spatial structure perspectives, achieving discriminative terrain feature aggregation. SATSNet is built around two essential modules: the Entropy-Guided Global Sparsification Module (EGGSM) and the Semantic-Spatial Fusion Module (SSFM). EGGSM performs selective channel-wise global semantic modeling through entropy-guided sparsification, providing representative long-range dependency semantic cues for guidance. SSFM comprises the Attention-Entropy Fusion Unit and the Dynamic Weight Allocation Unit, which leverage entropy-based attention to explore the semantic-spatial complementarity among adjacent features, facilitating the aggregation of semantic-aware terrain features. Validation on multiple challenging wild datasets, as well as large-scale street-view and real-world application datasets, demonstrates that SATSNet outperforms existing state-of-the-art methods in navigable area segmentation, providing a robust solution for safe autonomous navigation.

During the withering process of white tea, the degradation of astringent flavonol glycosides (FGs) plays a vital role in enhancing quality. However, the molecular mechanism underlying this process remains unclear. This study aimed to clarify the molecular mechanism and physiological significance of the degradation of nine key astringent FGs during white tea withering. Two key genes, CsGH3B (β-glucosidase) and CsPPO1 (polyphenol oxidase), were identified using FGs quantification, transcriptomics, and weighted gene co-expression network analysis (WGCNA). Prokaryotic expression, protein purification, and in vitro enzyme activity assays confirmed that recombinant CsGH3B hydrolyzed all nine FGs, whereas recombinant CsPPO1 did not. Transient inhibition of CsGH3B expression in tea leaves significantly increased the contents of the nine FGs (P < 0.05). Overexpression of CsGH3B in Nicotiana benthamiana promoted the hydrolysis of FGs, further confirming its in vivo function. Molecular docking revealed that CsGH3B binds to FGs through hydrogen bonds and hydrophobic interactions. Our results indicate that dehydration stress during white tea withering induces the accumulation of reactive oxygen species, which may be associated with the upregulation of CsGH3B, catalyzing the hydrolysis of FGs to generate potent antioxidant flavonol aglycones (quercetin and kaempferol) that potentially contribute to oxidative stress alleviation. This process also reduces the bitterness and astringency of white tea products by lowering FG content, thereby improving taste quality. Together, these findings provide a molecular-level insight into the concept that "adversity yields fine tea" and offer a potential theoretical basis for postharvest flavor regulation and quality improvement of white tea.

The presence of phenolic hydroxyl groups (Ph-OH) in lignin leads to generate abundant undesirable heavy components in pyrolysis oil, which hinders the application of pyrolysis oil as a fuel or a source of fine chemicals. In view of this, dimethyl sulfate ((CH3O)2SO2) was employed to selectively mask the Ph-OH of corncob alkali lignin (CAL) and softwood alkali lignin (SAL) to varying degrees. The intermittent pyrolysis results pronounced that methylated lignin pyrolysis produced more pyrolysis oil and less pyrolytic char compared to unmethylated lignin, due to the inhibition of polymerization reaction. However, the selectivity of total and S/G-monophenols gradually decreased with increasing degree of Ph-OH methylation. Furthermore, the Ph-OH methylation did not promote the formation of methoxyphenols, as evidenced by the decreasing trend of methoxyphenols selectivity with increasing Ph-OH methylation degree, which was ascribed to Ph-OH methylation promoting C−O bond cleavage during pyrolysis to form C-monophenols or aromatics. The two-dimensional heteronuclear single quantum coherent nuclear magnetic resonance (2D HSQC NMR) results indicated that Ph-OH masking was beneficial for the breakage of β−O−4 linkages, and the cleavage degree depended on the proportion of lignin structural units. The combination of gel permeation chromatography (GPC) and ultra-high performance liquid chromatography tandem quadrupole time-of-flight mass spectrometry (UHPLC/QTOF-HRMS) confirmed that Ph-OH masking reduced the weight average (Mw) and number average (Mn) molecular weight of pyrolysis oil, as well as inhibited the formation of heavy components.

The temperature sensitivity (Q10) of soil microbial respiration (Rs) is a critical parameter for predicting the response of microbially mediated decomposition of global soil organic carbon (SOC) to climate change. However, the variations in Q10 across horizontal and vertical spatial gradients remain contentious. In this study, we conducted a simulated soil warming incubation experiment across temperature gradients of 5 °C, 15 °C, 25 °C, and 35 °C using soils collected from the southern subtropical forest (SSF), mid-subtropical forest (MSF) and temperate forest (TF) in China. Soil samples were obtained from four depth intervals along a 60 cm soil profile: 0–15 cm, 15–30 cm, 30–45 cm, and 45–60 cm. We measured soil microbial respiration, SOC fractions, soil chemical properties, microbial community structure and activity. Q10 values were calculated, and the underlying mechanistic relationships among these variables were examined. Significant spatial variations in Q10 were observed (P < 0.05): (1) in the 30–60 cm soil layers, Q10 values in TF were significantly higher than those in SSF and MSF; (2) with in SSF, Q10 in the topsoil (0–15 cm) was markedly greater than that in the deep soil (45–60 cm). Horizontally, the higher Q10 values in TF appear to be influenced with higher-quality carbon substrates, a greater abundance of K-strategies microbial taxa, higher microbial activity and prolonged exposure to low temperatures. Vertically, in the SSF, the higher Q10 in topsoil was primarily attributed to higher SOC content, the presence of more labile carbon substrates and enhanced microbial activity. These findings underscore the important roles of carbon quality, microbial life-history strategies (K-strategies) and microbial activity in mediating the Q10 of Rs, especially in deep soils which are easily ignored. Based on these findings, it can be predicted that under global warming scenarios, temperate forests may experience accelerated SOC decomposition at horizontal spatial scales, especially in deeper soil layers. In subtropical forests, topsoil may exhibit more rapid carbon loss along vertical gradients.

Honey bees are economically important pollinators and widely studied model organisms. The phenomenon of honey bee caste differentiation manifested by ovary development has been a hot topic in apicultural research. Juvenile hormone (JH) has a critical role in caste differentiation in honey bees. Here, we analyzed the morphology and transcriptome changes in honey bee ovary after knocking down the expression of the JH answer gene AmKr-h1 by RNAi. As a result, the number of ovarian tubes was significantly reduced in the RNAi group and a total of 920 differentially expressed genes (DEGs) were found between the ovaries of the RNAi and negative control (NC) groups. Many of these genes are closely related to honey bee caste differentiation. In addition, 109 differentially expressed alternative splicing events (DEASEs) associated with 97 genes were identified between the RNAi group and NC group. The results of this study demonstrate that AmKr-h1 significantly influences ovarian caste differentiation and gene expression in honey bees. Our results offer gene expression data that deepen the understanding of the molecular mechanisms underlying ovarian differentiation between queens and workers.