Contrasting nitrogen transport patterns in subtropical basins revealed by combined multiple isotopic analyzes and hydrological simulations.

Although enhancing the knowledge of nitrogen (N) dynamics in aquatic systems is crucial for basin N management, there is still a lack of theories on the patterns of basin N sources and transport because of the intricate influence of human activities, climatic conditions, landscape patterns, and topography on the trajectory of basin N. To shed new light on the patterns of basin N sources and transport in the Chinese subtropical monsoon region, this study provides a comprehensive approach combining multiple isotopes and hydrological model based on monthly records of hydro-chemistry and isotopes (18O-NO3- /15N-NO-3 and 18O-H2O /2H-H2O) for river water, groundwater and rainfall in three basins over multiple years. Our observations of hydro-chemistry showed that fluvial N levels in highly urbanized basins (3.05 ± 1.42 mg·L-1) were the highest and were characterized by higher levels in the dry season. In the agricultural basin, fluvial N levels in February and March were approximately 1.9 times higher than those in the other months. The fluvial N load was higher in agricultural basins (0.624-0.728 T N km -2 y -1) than in urban basins (0.558 T N km -2 y -1), primarily because of variations in sewage treatment rates and fertilizer application. In highly urbanized basin, manure and sewage (46.9 %) were the dominant sources of fluvial N, which were discharged into rivers after treatment. In the plain agricultural basin, a substantial portion of diffused residential sewage leaches into aquifers and is stored. In the hilly agro-forest mixed basin, the high baseflow coefficient (75.8 %) and the key role of groundwater N, mainly from soil N (27.3 %), chemical fertilizers (20.2 %), manure and sewage (46.6 %), to fluvial N (26.5 %) indicated that a high proportion of the N sources leached into the aquifer and were then transported to rivers. For the first time, this study integrated multiple methods to substantiate the proposed typical patterns of N sources and transport within the basins. These findings have significant implications for tailored basin-specific N management strategies.

Two telomere-to-telomere gapless genomes reveal insights into Capsicum evolution and capsaicinoid biosynthesis.

Chili pepper (Capsicum) is known for its unique fruit pungency due to the presence of capsaicinoids. The evolutionary history of capsaicinoid biosynthesis and the mechanism of their tissue specificity remain obscure due to the lack of high-quality Capsicum genomes. Here, we report two telomere-to-telomere (T2T) gap-free genomes of C. annuum and its wild nonpungent relative C. rhomboideum to investigate the evolution of fruit pungency in chili peppers. We precisely delineate Capsicum centromeres, which lack high-copy tandem repeats but are extensively invaded by CRM retrotransposons. Through phylogenomic analyses, we estimate the evolutionary timing of capsaicinoid biosynthesis. We reveal disrupted coding and regulatory regions of key biosynthesis genes in nonpungent species. We also find conserved placenta-specific accessible chromatin regions, which likely allow for tissue-specific biosynthetic gene coregulation and capsaicinoid accumulation. These T2T genomic resources will accelerate chili pepper genetic improvement and help to understand Capsicum genome evolution.

GBDT-IL: Incremental Learning of Gradient Boosting Decision Trees to Detect Botnets in Internet of Things.

The rapid development of the Internet of Things (IoT) has brought many conveniences to our daily life. However, it has also introduced various security risks that need to be addressed. The proliferation of IoT botnets is one of these risks. Most of researchers have had some success in IoT botnet detection using artificial intelligence (AI). However, they have not considered the impact of dynamic network data streams on the models in real-world environments. Over time, existing detection models struggle to cope with evolving botnets. To address this challenge, we propose an incremental learning approach based on Gradient Boosting Decision Trees (GBDT), called GBDT-IL, for detecting botnet traffic in IoT environments. It improves the robustness of the framework by adapting to dynamic IoT data using incremental learning. Additionally, it incorporates an enhanced Fisher Score feature selection algorithm, which enables the model to achieve a high accuracy even with a smaller set of optimal features, thereby reducing the system resources required for model training. To evaluate the effectiveness of our approach, we conducted experiments on the BoT-IoT, N-BaIoT, MedBIoT, and MQTTSet datasets. We compared our method with similar feature selection algorithms and existing concept drift detection algorithms. The experimental results demonstrated that our method achieved an average accuracy of 99.81% using only 25 features, outperforming similar feature selection algorithms. Furthermore, our method achieved an average accuracy of 96.88% in the presence of different types of drifting data, which is 2.98% higher than the best available concept drift detection algorithms, while maintaining a low average false positive rate of 3.02%.