Abstract
Protein structure prediction is critical to functional annotation of the massively accumulated biological sequences, which prompts an imperative need for the development of high-throughput technologies. As a first and key step in protein structure prediction, protein structural class prediction becomes an increasingly challenging task. Amongst most homological-based approaches, the accuracies of protein structural class prediction are sufficiently high for high similarity datasets, but still far from being satisfactory for low similarity datasets, i.e., below 40% in pairwise sequence similarity. Therefore, we present a novel method for accurate and reliable protein structural class prediction for both high and low similarity datasets. This method is based on Support Vector Machine (SVM) in conjunction with integrated features from position-specific score matrix (PSSM), PROFEAT and Gene Ontology (GO). A feature selection approach, SVM-RFE, is also used to rank the integrated feature vectors through recursively removing the feature with the lowest ranking score. The definitive top features selected by SVM-RFE are input into the SVM engines to predict the structural class of a query protein. To validate our method, jackknife tests were applied to seven widely used benchmark datasets, reaching overall accuracies between 84.61% and 99.79%, which are significantly higher than those achieved by state-of-the-art tools. These results suggest that our method could serve as an accurate and cost-effective alternative to existing methods in protein structural classification, especially for low similarity datasets.
Highlights
As the basic compositions of life, proteins play a central role in most cellular functions such as gene regulation, metabolism and cell proliferation
We propose a novel computation method that combines Support Vector Machine (SVM) with PSI-BLAST profile, physical-chemical property and functional annotations to further improve the prediction of protein structural class
Parameter selection In this study, we used a grid search strategy to select the parameters in LIBSVM, which depend on the dimension Dim of the top feature vector of proteins
Summary
As the basic compositions of life, proteins play a central role in most cellular functions such as gene regulation, metabolism and cell proliferation. In order to interpret the function of a new protein sequence, it is fundamental to understand its 3D structure. Since the knowledge of protein structural class provides useful information towards the determination of its 3D structure, prediction of protein structural class from sequence data becomes a hot topic in computational biology, especially with the development of high-throughput technologies [1]. Proteins have irregular surfaces and complex 3D structures, but they are formed regularly in regional fold patterns at secondary structure level. Based on the contents of their secondary structures, known protein structures are classified into four categories, all-a, all-b, a/b and a+b. All-a and all-b proteins consist of only ahelices and b-strands, respectively. Experimental approaches to determining the structure information of a protein, including X-ray Diffraction and Nuclear Magnetic Resonance, are costly and time-consuming, and not capable of completely meeting researchers’ demands. Highthroughput computational approaches are brought to the forefront of this issue
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
