Xuzhou Institute of Technology

About: Xuzhou Institute of Technology is a education organization based out in Xuzhou, China. It is known for research contribution in the topics: Catalysis & Computer science. The organization has 1696 authors who have published 1521 publications receiving 13541 citations.

FCGCNMDA: predicting miRNA-disease associations by applying fully connected graph convolutional networks.

Abstract: Growing evidence indicates that the development and progression of multiple complex diseases are influenced by microRNA (miRNA). Identifying more miRNAs as biomarkers for clinical diagnosis, treatment and prognosis is vital to promote the development of bioinformatics and medicine. Considering that the traditional biological experimental methods are generally time-consuming and expensive, high-efficient computational methods are encouraged to uncover potential disease-related miRNAs. In this paper, FCGCNMDA is presented to predict latent miRNA-disease associations by utilizing fully connected graph convolutional networks. Specially, our method first constructs a fully connected graph in which edge weights represent correlation coefficient between any two pairs of miRNA-disease pair, and then feeds this fully connected graph along with miRNA-disease pairs feature matrix into a two-layer graph convolutional networks (GCN) for training. At last, we utilize the trained network to predict the scores for unknown miRNA-disease pairs. As a result, FCGCNMDA achieves AUC value of [Formula: see text] and AUPRC value of [Formula: see text] in HMDD v2.0 based on five-fold cross validation. Moreover, case studies on Lymphoma, Breast Neoplasms and Prostate Neoplasms shown that 98%, 98%, 98% of the top 50 selected miRNAs were validated by recent experimental evidence. From above results, we can deduce that FCGCNMDA can be regarded as reliable method for potential miRNA-disease associations prediction.

Influence of BaTiO 3 Nanoparticle Addition on Microstructure and Mechanical Properties of Sn-58Bi Solder

Abstract: The effect of BaTiO3 nanoparticles on the microstructure, wettability, and mechanical properties of Sn-58Bi solder has been investigated. The results show that the microstructure of the Sn-58Bi-x wt.%BaTiO3 (x = 0.5, 1, 2, 3) composite solders is refined and homogeneous due to increasing heterogeneous nucleation sites of BaTiO3 nanoparticles in the solder matrix. The best wettability with spreading coefficient of 0.86 was obtained for Sn-Bi-1 wt.%BaTiO3, increased by 10.24% compared with Sn-58Bi solder. High mechanical properties were achieved for Sn-58Bi solder with small addition (0.5 wt.% to 3 wt.%) of BaTiO3. The ultimate tensile strength (UTS) of Sn-58Bi is 44.7 MPa, and that of Sn-Bi-1%BaTiO3 is 59.1 MPa. Fracture surface analysis indicated that the microscopic fracture of Sn-Bi-1%BaTiO3 solder is composed of tearing marks and ductility is improved. Therefore, Sn-Bi-1%BaTiO3 composite solder is one of the candidates for packaging and interconnection applications.