The performance of regional groundwater level (GWL) prediction model hinges on understanding intricate spatiotemporal correlations among monitoring wells. In this study, a graph convolutional network (GCN) with a long short-term memory (LSTM) (GCN–LSTM) model is introduced for GWL prediction utilizing data from 16 wells located in the northeastern Xiangtan City, China. This model is designed to account for both the hybrid temporal dependencies and spatial autocorrelations among wells. It consists of two parts: the spatial part employs GCNs to extract spatial characteristics from a spatial self-similarity weight matrix and an attribute self-similarity weight matrix among wells; the temporal part utilizes a LSTM module to capture the temporal patterns of GWL sequences, along with monthly precipitation and temperature data. This model dynamically predicts changes in groundwater levels, achieving higher accuracy on average compared to single-well predictions using LSTM. By incorporating both temporal dependencies and spatial autocorrelations, the GCN–LSTM model demonstrated an average improvement in goodness-of-fit of approximately 11.21% over the LSTM-based model for individual wells. Its application holds significant reference value for the sustainable utilization and development of groundwater resources in Xiangtan City.
The radial migration of cortical pyramidal neurons (PNs) during corticogenesis is necessary for establishing a multilayered cerebral cortex. Neuronal migration defects are considered a critical etiology of neurodevelopmental disorders, including autism spectrum disorders (ASDs), schizophrenia, epilepsy, and intellectual disability (ID). TRIO is a high-risk candidate gene for ASDs and ID. However, its role in embryonic radial migration and the etiology of ASDs and ID are not fully understood. In this study, we found that the in vivo conditional knockout or in utero knockout of Trio in excitatory precursors in the neocortex caused aberrant polarity and halted the migration of late-born PNs. Further investigation of the underlying mechanism revealed that the interaction of the Trio N-terminal SH3 domain with Myosin X mediated the adherence of migrating neurons to radial glial fibers through regulating the membrane location of neuronal cadherin (N-cadherin). Also, independent or synergistic overexpression of RAC1 and RHOA showed different phenotypic recoveries of the abnormal neuronal migration by affecting the morphological transition and/or the glial fiber-dependent locomotion. Taken together, our findings clarify a novel mechanism of Trio in regulating N-cadherin cell surface expression via the interaction of Myosin X with its N-terminal SH3 domain. These results suggest the vital roles of the guanine nucleotide exchange factor 1 (GEF1) and GEF2 domains in regulating radial migration by activating their Rho GTPase effectors in both distinct and cooperative manners, which might be associated with the abnormal phenotypes in neurodevelopmental disorders.
CAN bus communication system was established mainly using SimEvents module in the software of Matlab/Simulink. Performance indexes of CAN bus system, which embraces worst-case response time of messages and bus load, were separately gained by theoretical calculation, experiments in Matlab and CANoe. The calculation and experiment results were compared sufficiently, and it concludes that CAN bus system established in Matlab/Simulink can efficiently represent the message transfer sequences on the CAN bus, and achieve an exact result of the performance analysis of the CAN bus system. With CAN bus system established in Matlab/Simulink, test modules for ECU (Electric Control Unit) testing can conveniently be created, and it is an efficient testing tool for CAN bus system design.