Prediction of protein structural classes by different feature expressions based on 2-D wavelet denoising and fusion

Abstract

BackgroundProtein structural class predicting is a heavily researched subject in bioinformatics that plays a vital role in protein functional analysis, protein folding recognition, rational drug design and other related fields. However, when traditional feature expression methods are adopted, the features usually contain considerable redundant information, which leads to a very low recognition rate of protein structural classes.ResultsWe constructed a prediction model based on wavelet denoising using different feature expression methods. A new fusion idea, first fuse and then denoise, is proposed in this article. Two types of pseudo amino acid compositions are utilized to distill feature vectors. Then, a two-dimensional (2-D) wavelet denoising algorithm is used to remove the redundant information from two extracted feature vectors. The two feature vectors based on parallel 2-D wavelet denoising are fused, which is known as PWD-FU-PseAAC. The related source codes are available at https://github.com/Xiaoheng-Wang12/Wang-xiaoheng/tree/master.ConclusionsExperimental verification of three low-similarity datasets suggests that the proposed model achieves notably good results as regarding the prediction of protein structural classes.

A Wavelet-Based Learning Model Enhances Molecular Prognosis in Pancreatic Adenocarcinoma

“…From the perspective of digital signal processing, S. Wang and X. Wang introduced a two-dimensional wavelet into protein structure prediction for denoising [9]. Lin et al used stationary wavelet transform for sequence similarity analysis, where wavelet transform was used to convert complex numbers obtained from cluster mapping into feature vectors [10].…”

Prediction of protein structural class based on symmetrical recurrence quantification analysis

“…From Table 6, it is clearly shown that our best configuration i.e. RF[8-SRQA-I ] outperformed the recent results of Wang et al [39]. In addition, from Tables A.7-A.9 we see that RF[ 8-SRQA-I], RF[8-SRQA-R] overpass the recent results of Wang et al [39] for any type of symmetry.…”

“…In Table 6 a comparison is made between our best results obtained with RF[8-SRQA-I] and other existing methods [20,38,26,39,40,41]. From Table 6, it is clearly shown that our best configuration i.e.…”

Graph-Based Analysis of RNA Secondary Structure Similarity Comparison

“…A combination of wavelet transform and neural network has been proposed to achieve highly accurate machine fault diagnosis [41]. In [42], wavelet analysis was used to identify membrane protein types. It was also used to realize RNA secondary structure similarity analysis.…”