No abstract

It is tough to detect unexpected drug–drug interactions (DDIs) in poly-drug treatments because of high costs and clinical limitations. Computational approaches, such as deep learning-based approaches, are promising to screen potential DDIs among numerous drug pairs. Nevertheless, existing approaches neglect the asymmetric roles of two drugs in interaction. Such an asymmetry is crucial to poly-drug treatments since it determines drug priority in co-prescription. This paper designs a directed graph attention network (DGAT-DDI) to predict asymmetric DDIs. First, its encoder learns the embeddings of the source role, the target role and the self-roles of a drug. The source role embedding represents how a drug influences other drugs in DDIs. In contrast, the target role embedding represents how it is influenced by others. The self-role embedding encodes its chemical structure in a role-specific manner. Besides, two role-specific items, aggressiveness and impressionability, capture how the number of interaction partners of a drug affects its interaction tendency. Furthermore, the predictor of DGAT-DDI discriminates direction-specific interactions by the combination between two proximities and the above two role-specific items. The proximities measure the similarity between source/target embeddings and self-role embeddings. In the designated experiments, the comparison with state-of-the-art deep learning models demonstrates the superiority of DGAT-DDI across a direction-specific predicting task and a direction-blinded predicting task. An ablation study reveals how well each component of DGAT-DDI contributes to its ability. Moreover, a case study of finding novel DDIs confirms its practical ability, where 7 out of the top 10 candidates are validated in DrugBank.

No abstract

An experiment was performed to investigate the influence of four dissolved salts (NaCl, CaCl 2 , Na 2 SO 4 and MgSO 4 ) on heat transfer. Concentration of all salts varied from 0.01 M to 0.4 M. And deionized water (DI water) was used as the solvent. The aluminum alloy 2024 thin sheets were heated to 495℃ and quenched by two spray nozzles with a constant range of volumetric flux. Time-temperature curves were smoothed by B-spline approximation with a smoothing parameter of 10 -4 . The results indicate that only MgSO 4 solution increases both nucleate boiling and transition boiling heat flux. Solution of CaCl 2 has the greatest influence on nucleate boiling heat transfer, followed by Na 2 SO 4 compared with the minimum influence of NaCl. These effects are attributed to the surface tension gradient, transition concentration and vapor pressure for different salts. The largest heat flux enhancement is observed at concentration of 0.2 M for solutions of NaCl, CaCl 2 and MgSO 4 in contrast to 0.06 M for Na 2 SO 4 .

In complex networks, it is significant to rank the nodes according to their importance. In this paper we present an algorithm based on an improved Structural Holes method to identify the key nodes of a complex network. Since our approach does not need to consider the global structure of a network but only consider the number of one node's neighbors and it's next nearest neighbors, the nodes importance can be calculated with local information of a complex network. Experimental results of ARPA net show that our method is better than some important ranking measures such as betweenness, degree or closeness. It is very useful for evaluating the key nodes in large scale and complicated networks, in which evaluation of nodes importance is almost impossible to calculate with global information.