Search

Over the past few years, there has been significant research on applying Transformer models to time series prediction, yielding promising results. Simultaneously, researchers have begun exploring the utilization of Transformers for traffic prediction in order to mitigate the nonlinear spatial–temporal correlation inherent in traffic data. Some of these studies have attempted to characterize spatial–temporal features by incorporating embedding structures, with the goal of improving performance of the model. However, existing methods have not adequately addressed the issue of spatial–temporal correlation. To address these limitations, we propose the Transformer Network with Decoupled Spatial–Temporal Embedding (DSTET) model for traffic flow prediction. The key aspect of our model is its ability to effectively decouple the spatial and temporal embedding through the implementation of the Decoupled Spatial–Temporal Embedding structure. This structure enhances the characterization of spatial–temporal features, ultimately improving the performance of traffic prediction based on the Transformer model. Through experiments conducted on six real-world traffic datasets, our model consistently outperforms multiple baseline models, demonstrating its capability to address the identified problems. Moreover, we substantiate the efficacy of the suggested components via ablation experiments and furnish a thorough analysis of the attention weight matrix to clarify the functioning of the model. [ABSTRACT FROM AUTHOR]

Structure-based drug design depends on the detailed knowledge of the three-dimensional (3D) structures of protein–ligand binding complexes, but accurate prediction of ligand-binding poses is still a major challenge for molecular docking due to deficiency of scoring functions (SFs) and ignorance of protein flexibility upon ligand binding. In this study, based on a cross-docking dataset dedicatedly constructed from the PDBbind database, we developed several XGBoost-trained classifiers to discriminate the near-native binding poses from decoys, and systematically assessed their performance with/without the involvement of the cross-docked poses in the training/test sets. The calculation results illustrate that using Extended Connectivity Interaction Features (ECIF), Vina energy terms and docking pose ranks as the features can achieve the best performance, according to the validation through the random splitting or refined-core splitting and the testing on the re-docked or cross-docked poses. Besides, it is found that, despite the significant decrease of the performance for the threefold clustered cross-validation, the inclusion of the Vina energy terms can effectively ensure the lower limit of the performance of the models and thus improve their generalization capability. Furthermore, our calculation results also highlight the importance of the incorporation of the cross-docked poses into the training of the SFs with wide application domain and high robustness for binding pose prediction. The source code and the newly-developed cross-docking datasets can be freely available at https://github.com/sc8668/ml%5fpose%5fprediction and https://zenodo.org/record/5525936, respectively, under an open-source license. We believe that our study may provide valuable guidance for the development and assessment of new machine learning-based SFs (MLSFs) for the predictions of protein–ligand binding poses. [ABSTRACT FROM AUTHOR]

The study of multi-dimensional expansion of urban space (MEUS) addresses the laws of urban spatial expansion from all directions and angles. Using Nanjing as an example, this paper constructs multi-temporal, urban three-dimensional models based on RS and GIS technology and then conducts qualitative and quantitative analysis of MEUS using plot ratio change maps and MEUS quantitative index for built-up areas. Based especially on the concept of volume growth contribution rate, this paper analyzes the characteristics of MEUS in different stages. The results show that in 2000–2004, planar expansion played the main role, the internal potential development (IPD) intensity of the urban built-up areas was relatively large, and the volume growth contribution rate was low; in 2004–2008, planar expansion accelerated, and IPD slowed down; in 2008–2012, planar expansion slowed while IPD intensity increased; the contribution rates of volume growth of urban IPD for the three periods were 22.21%, 24.51% and 73.38%, respectively. This study expands the research perspective of urban spatial expansion, and the adopted methods are instructive and meaningful for MEUS research. In addition, the results of this study will deepen the understanding of MEUS laws and help improve scientific decision-making for urban planning and urban land use management. [ABSTRACT FROM AUTHOR]

In this paper, the artificial neural network (ANN) model was used to evaluate the degree of intensive urban land use in Nanjing City, China. The construction and application of the ANN model took into account the comprehensive, spatial and complex nature of urban land use. Through a preliminary calculation of the degree of intensive land use of the sample area, representative sample area selection and using the back propagation neural network model to train, the intensive land use level of each evaluation unit is finally determined in the study area. Results show that the method can effectively correct the errors caused by the limitations of the model itself and the determination of the ideal value and weights when the multifactor comprehensive evaluation is used alone. The ANN model can make the evaluation results more objective and practical. The evaluation results show a tendency of decreasing land use intensity from the core urban area to the periphery and the industrial functional area has relatively low land use intensity compared with other functional areas. Based on the evaluation results, some suggestions are put forward, such as transforming the mode of urban spatial expansion, strengthening the integration and potential exploitation of the land in the urban built-up area, and strengthening the control of the construction intensity of protected areas. [ABSTRACT FROM AUTHOR]

The Leishan-Rongjiang antimony ore field (LAOF) is in a unique geotectonic location in the uplift between the Youjiang and Xiangzhong basins. This paper focuses on two representative deposits in the LAOF: the Bameng and Peize antimony (Sb) deposits. We analyzed fluid inclusions (FIs) in stibnite and coexisting quartz, as well as the sulfur isotopic composition of stibnite, to better understand the nature of the ore-forming fluid and the metallogenic process. The FIs data from samples of the stibnite and coexisting quartz indicate that the ore-forming fluids were characterized by low-temperature (150-210 °C), low-salinity (1.5 wt%-6.0 wt% NaCl equiv.), and low-density (0.872-0.961 g/cm). The δS values of stibnite (−8.21‰ to 3.76‰, average = −6.30‰) fall in between the sulfur isotopic compositions of the mantle and of biogenic sulfur in sedimentary rocks. However, the δS values (−4.41‰ to +0.04‰, average = −2.49‰) of the ore-forming fluids are generally closer to the sulfur isotopic composition of the mantle source, indicating that the sulfur in the LAOF was mainly sourced from the mantle, but with possible involvement of biogenic sulfur. In addition, FIs petrography and ore deposit geology show that fluid boiling resulted from an abrupt decrease in pressure, which may have triggered the precipitation of stibnite. We conclude that low-temperature, dilute hydrothermal fluids with mixed origins migrated along the regional fault and interacted with the wall rock, extracting the ore-forming materials. Then, the ore-forming fluids were injected into the fault fracture zones. [ABSTRACT FROM AUTHOR]

Iodine was relatively enriched in the sedimentary phosphorite that formed during the Doushantuo period of Late Ediacaran, South China, especially in the Central of Guizhou Province, Southwest China. The resources of the associated iodine in the phosphorite are extremely high. The analysis and test of elements such as iodine in the phosphorus-bearing rock series reveal that iodine occurs only in the phosphorite and the iodine has positive correlation with PO and CaO which are the main components of apatite, while has negative correlation with major oxides including MgO, NaO, FeO, TiO and KO, which showed that the iodine is mainly rich in phosphorite. Meanwhile, it was observed that the iodine content is the highest in the gray-white clastic phosphorite, but is relatively low in the gray-black carbonaceous phosphorite. Discussion of the sedimentary environment in the research area showed that the gray, gray-white nest-like, particle-like, and fragmental algae-enriched phosphorite feature the most abundant iodine element, with the average content of 59 mg kg and the maximum value of 98 mg kg. The compacted blocky phosphorite deposited in the neritic marine basin facies only has trace iodine, maybe due to the impact on the iodine absorption caused by the inhibition of algae propagation under anaerobic and reduction conditions. Results of correlation analysis illustrate that the relationship between iodine and the number of algae was suggested. Based on the aforementioned regulations and conditions for the iodine concentration in the phosphorite of the Late Edicanran Doushantuoian, the main iodine-enriched of ore, ore type, and facies belt were considered biogenic apatite, clastic phosphorite, and neritic marine platform facies, respectively. [ABSTRACT FROM AUTHOR]

In the study of the phosphate-bearing stratum at the bottom of the Cambrian system, the authors found that there occurred carbon isotope negative anomalies in the Taozichong section phosphate-bearing stratum in Qingzhen, Guizhou Province; they can be correspondingly compared with other synsedimentary carbon isotope negative anomalies both at home and abroad. The results showed that there occurred three negative anomalies of carbon isotopes in the Neoproterozoic-Early Cambrian Taozichong Formation, indicating that the temporal palaeo-oceanographical environment changed significantly, in which there happened two times of intensive carbon isotope variation, corresponding to creature extinction. Meanwhile, it is believed that the carbon isotope negative anomalies in phosphorites were caused by the ascending water mass of ocean current with the negative carbon isotopic composition of deep-ocean hydrothermal deposits. [ABSTRACT FROM AUTHOR]

The evolution of the global sulfur isotope curve was plotted based on the δ 34S values of evaporates resultant from oceanic evaporation. In the long period of geological history the δ 34S values showed obvious peaks for three times during the process of ancient oceans’ sulfur isotope evolution, namely the Early Cambrian (+30‰), the Late Devonian (+25‰) and the Permian-Triassic transition interval (+17‰), but the causes of the abnormal rise of sulfur isotopic values during the geological period are still in question. In this paper, 18 samples collected from a large Devonian barite deposit from Zhenning County were analyzed to determine their δ 34S values, revealing that the 18 samples have very high δ 34S values (δ 34S=41.88‰–+68.39‰), with an average close to 56.30‰, which are higher than the isotopic values of contemporary sulfates (+17‰–+25‰). A comparative analysis was conducted of the emerging of high δ 34S barite deposits (from Cambrian and Devonian) and the δ 34S variation curves of the ancient oceans. The results indicate that the time when the obvious peaks of δ 34S values appeared and the time of massive sedimentation of high δ 34S barite deposits are very close to each other, which, in our opinion, is not a coincidence. There may exist some correlations between the sulfur isotope evolution of ancient oceans during the diverse periods of geological history and the massive sedimentation of high δ 34S barite deposits. Therefore, it is inferred that perhaps it was the massive sedimentation of high δ 34S barites that caused the sharp rise of δ 34S values in a short period of time. [ABSTRACT FROM AUTHOR]