The recognition of protein folds is an important step in the prediction of protein structure and function. Recently, an increasing number of researchers have sought to improve the methods for protein fold recognition. Following the construction of a dataset consisting of 27 protein fold classes by Ding and Dubchak in 2001, prediction algorithms, parameters and the construction of new datasets have improved for the prediction of protein folds. In this study, we reorganized a dataset consisting of 76-fold classes constructed by Liu et al. and used the values of the increment of diversity, average chemical shifts of secondary structure elements and secondary structure motifs as feature parameters in the recognition of multi-class protein folds. With the combined feature vector as the input parameter for the Random Forests algorithm and ensemble classification strategy, we propose a novel method to identify the 76 protein fold classes. The overall accuracy of the test dataset using an independent test was 66.69%; when the training and test sets were combined, with 5-fold cross-validation, the overall accuracy was 73.43%. This method was further used to predict the test dataset and the corresponding structural classification of the first 27-protein fold class dataset, resulting in overall accuracies of 79.66% and 93.40%, respectively. Moreover, when the training set and test sets were combined, the accuracy using 5-fold cross-validation was 81.21%. Additionally, this approach resulted in improved prediction results using the 27-protein fold class dataset constructed by Ding and Dubchak.

In order to predict enzyme subclasses, this paper builds a new enzyme database in term of previous ideas and methods. Based on protein sequence, by selecting increment of diversity value, low-frequency of power spectral density, matrix scoring values and motif frequency as characteristic parameters to describe the sequence information, a Random Forest algorithm for predicting enzyme subclass is proposed. Using the Jack-knife test, the overall success rate identifying the 18 subclasses of oxidoreductases, the 8 subclasses of transferases, the 5 subclasses of hydrolases, the 6 subclasses of lyases, the 6 subclasses of isomerases, and the 6 subclasses of ligases are 90.86%, 95.24%, 96.42%, 98.60%, 97.53% and 98.03%. Furthermore, the same way is used to the previous database, the better results are obtained.

With the rapid development of silicon photonic chips and integrated photonic circuits, erbium-doped optical waveguide amplifiers have been received more and more attention in order to compensate for the transmission and coupling losses caused by the integration of optical devices on a chip. Pumping wavelength and pumping efficiency directly affect the gain and noise figure of the amplifier. In this paper, we propose an innovative dual-wavelength pumping method based on an erbium-ytterbium co-doped optical waveguide amplifier with simultaneous pumping at 980 nm and 1480 nm. A relaxation method based on the fourth-order Range-Kutta method is used to solve the rate and propagation equations and simulate the gain characteristics of the dual-wavelength pumping method are simulated for different ytterbium-erbium ion concentration ratios, erbium ion concentrations and ratio K between the 980 nm and total pump power. From the simulation results, it can be seen that the gain of the dual-wavelength pumping method is higher than that of the single-wavelength pumping methods when the erbium ion concentration exceeds 3×1026 /m3. At higher erbium ion concentrations, the dual-wavelength pumping method can provide higher gain for optical waveguide amplifiers, and may be able to become a new choice of pumping method for optical waveguide amplifiers.

No abstract