SGAE-MDA: Exploring the MiRNA-disease associations in herbal medicines based on semi-supervised graph autoencoder.

Research indicates that miRNAs present in herbal medicines are crucial for identifying disease markers, advancing gene therapy, facilitating drug delivery, and so on. These miRNAs maintain stability in the extracellular environment, making them viable tools for disease diagnosis. They can withstand the digestive processes in the gastrointestinal tract, positioning them as potential carriers for specific oral drug delivery. By engineering plants to generate effective, non-toxic miRNA interference sequences, it's possible to broaden their applicability, including the treatment of diseases such as hepatitis C. Consequently, delving into the miRNA-disease associations (MDAs) within herbal medicines holds immense promise for diagnosing and addressing miRNA-related diseases. In our research, we propose the SGAE-MDA model, which harnesses the strengths of a graph autoencoder (GAE) combined with a semi-supervised approach to uncover potential MDAs in herbal medicines more effectively. Leveraging the GAE framework, the SGAE-MDA model exactly integrates the inherent feature vectors of miRNAs and disease nodes with the regulatory data in the miRNA-disease network. Additionally, the proposed semi-supervised learning approach randomly hides the partial structure of the miRNA-disease network, subsequently reconstructing them within the GAE framework. This technique effectively minimizes network noise interference. Through comparison against other leading deep learning models, the results consistently highlighted the superior performance of the proposed SGAE-MDA model. Our code and dataset can be available at: https://github.com/22n9n23/SGAE-MDA.

Multi-Agent Deep Reinforcement Learning Based Dynamic Task Offloading in a Device-to-Device Mobile-Edge Computing Network to Minimize Average Task Delay with Deadline Constraints.

Device-to-device (D2D) is a pivotal technology in the next generation of communication, allowing for direct task offloading between mobile devices (MDs) to improve the efficient utilization of idle resources. This paper proposes a novel algorithm for dynamic task offloading between the active MDs and the idle MDs in a D2D-MEC (mobile edge computing) system by deploying multi-agent deep reinforcement learning (DRL) to minimize the long-term average delay of delay-sensitive tasks under deadline constraints. Our core innovation is a dynamic partitioning scheme for idle and active devices in the D2D-MEC system, accounting for stochastic task arrivals and multi-time-slot task execution, which has been insufficiently explored in the existing literature. We adopt a queue-based system to formulate a dynamic task offloading optimization problem. To address the challenges of large action space and the coupling of actions across time slots, we model the problem as a Markov decision process (MDP) and perform multi-agent DRL through multi-agent proximal policy optimization (MAPPO). We employ a centralized training with decentralized execution (CTDE) framework to enable each MD to make offloading decisions solely based on its local system state. Extensive simulations demonstrate the efficiency and fast convergence of our algorithm. In comparison to the existing sub-optimal results deploying single-agent DRL, our algorithm reduces the average task completion delay by 11.0% and the ratio of dropped tasks by 17.0%. Our proposed algorithm is particularly pertinent to sensor networks, where mobile devices equipped with sensors generate a substantial volume of data that requires timely processing to ensure quality of experience (QoE) and meet the service-level agreements (SLAs) of delay-sensitive applications.

Magnetic properties and rare earth element diffusion behavior of hot-deformed nanocrystalline dual-main-phase Nd-Ce-Fe-B magnets.

To inhibit the magnetic dilution effect of Ce in Nd-Ce-Fe-B magnets, a dual-alloy method is employed to prepare hot-deformed dual-main-phase (DMP) magnets using mixed nanocrystalline Nd-Fe-B and Ce-Fe-B powders. A REFe2 (1 : 2, where RE is a rare earth element) phase can only be detected when the Ce-Fe-B content exceeds 30 wt%. The lattice parameters of the RE2Fe14B (2 : 14 : 1) phase exhibit non-linear variation with the increasing Ce-Fe-B content due to the mixed valence states of Ce ions. Owning to inferior intrinsic properties of Ce2Fe14B compared to Nd2Fe14B, the magnetic properties of DMP Nd-Ce-Fe-B magnets almost decrease with the increase of Ce-Fe-B addition, but interestingly, the magnet with 10 wt% Ce-Fe-B addition exhibits an abnormally increased intrinsic coercivity H cj of 1215 kA m-1, together with the higher temperature coefficients of remanence (α = -0.110%/K) and coercivity (ß = -0.544%/K) in the temperature range of 300-400 K than the single-main-phase (SMP) Nd-Fe-B magnet with H cj = 1158 kA m-1, α = -0.117%/K and ß = -0.570%/K. The reason may be partly attributed to the increase of Ce3+ ions. Different from the Nd-Fe-B powders, the Ce-Fe-B powders in the magnet are difficult to deform into a platelet-like shape because of the lack of low melting point RE-rich phase due to the precipitation of the 1 : 2 phase. The inter-diffusion behavior between the Nd-rich region and Ce-rich region in the DMP magnets has been investigated by microstructure analysis. The significant diffusion of Nd and Ce into Ce-rich and Nd-rich grain boundary phases, respectively, was demonstrated. At the same time, Ce prefers to stay in the surface layer of Nd-based 2 : 14 : 1 grains, but less Nd diffuses into Ce-based 2 : 14 : 1 grains due to the 1 : 2 phase presented in the Ce-rich region. The modification of the Ce-rich grain boundary phase by Nd diffusion and the distribution of Nd in the Ce-rich 2 : 14 : 1 phase are beneficial for magnetic properties.

Inhomogeneity of the Backward Extruded NdFeB Ring Magnet Prepared from Amorphous Powders.

Back extrusion is an important process to prepare radially oriented NdFeB ring magnets. In this work, we fabricate the ring magnets using amorphous magnetic powders as the raw material. The microstructure, magnetic properties, corrosion resistance, and mechanical properties of the backward extruded magnet at different positions along the axial direction have been investigated, and the inhomogeneity of the magnet is clarified. The results showed that the grains in the middle region of the ring magnet exhibit a strong c-axis orientation, whereas the grains at the bottom and top regions are disordered with random orientation. The microstructure variation is related to the distribution of the grain boundary phase and the degree of grain deformation. Due to the microstructure difference, the magnetic properties, temperature stability, corrosion resistance, and mechanical properties in the middle region of the magnet are higher than those in the top and bottom regions. The exchange coupling between grains also varies in different regions, which is related to the grain size and grain boundary thickness. In addition, different Co element segregations were observed in different regions, which has a crucial effect on the Curie temperature and thermal stability of the magnet. The microstructure difference also leads to the variation of corrosion resistance and mechanical properties for the samples from different regions of the magnet. This work suggests that the amorphous powder can be used to directly prepare radially oriented ring magnets, and the inhomogeneity of the magnet should be fully understood.