No abstract

No abstract

Structural class characterizes the overall folding type of a protein or its domain. Most of the existing methods for determining the structural class of a protein are based on a group of features that only possesses a kind of discriminative information for the prediction of protein structure class. However, different types of discriminative information associated with primary sequence have been completely missed, which undoubtedly has reduced the success rate of prediction. We present a novel method for the prediction of protein structure class by coupling the improved genetic algorithm (GA) with the support vector machine (SVM). This improved GA was applied to the selection of an optimized feature subset and the optimization of SVM parameters. Jackknife tests on the working datasets indicated that the prediction accuracies for the different classes were in the range of 97.8-100% with an overall accuracy of 99.5%. The results indicate that the approach has a high potential to become a useful tool in bioinformatics.

No abstract

As the development of Wireless Sensor Network (WSN), software testing for WSN-based applications be-comes more and more important. Simulation testing is an important approach to WSN-based software testing, and TOSSIM is the most widely used simulation testing tool targeted at TinyOS which is the most popular operating system nowadays. However, simulation testing tools such as TOSSIM can not reveal program er-rors about communication detail or timing, and lack accurate power consumption model and even can not support power consumption estimation. In this paper, a hybrid testbed H-TOSSIM is proposed, which ex-tends TOSSIM with physical nodes. H-TOSSIM uses three physical nodes, of which, one shares the simu-lated environment with all virtual nodes to test the WSN program, and the other two bridge the real world and the simulated environment. H-TOSSIM combines the advantages of both the simulation in physical node and the simulation testing tools in WSN software testing. Through experiments, we show that H-TOSSIM really reveals program errors which the pure simulation testing can not capture, and can support power con-sumption estimation for large WSN with high accuracy and low hardware cost