Background: RNA-binding proteins establish posttranscriptional gene regulation by coordinating maturation, editing, transport, stability, and translation of cellular RNAs. Immunoprecipitation experiments could identify the interaction between RNA and proteins, but they are limited due to the experimental environment and material. Therefore, it is essential to construct computational models to identify the function sites. Objective: Although some computational methods have been proposed to predict RNA binding sites, the accuracy could be further improved. Moreover, it is necessary to construct a dataset with more samples to design a reliable model. Here we present a computational model based on multi-information sources to identify RNA binding sites. Methods: We construct an accurate computational model named CSBPI_Site, based on extreme gradient boosting. The specifically designed 15-dimensional feature vector captures four types of information (chemical shift, chemical bond, chemical properties and position information). Results: The satisfied accuracy of 0.86 and AUC of 0.89 were obtained by leave-one-out crossvalidation. Meanwhile, the accuracies were slightly different (range from 0.83 to 0.85) among the three classifiers algorithm, which showed that the novel features are stable and fit to multiple classifiers. These results showed that the proposed method is effective and robust for the identification of noncoding RNA binding sites. Conclusion: Our method based on multi-information sources is effective to represent the binding sites information among ncRNAs. The satisfied prediction results of Diels-Alder riboz-yme based on CSBPI_Site indicates that our model is valuable to identify the function site.
11 February 2019 in SAR and QSAR in Environmental Research
In Gram-negative bacteria, a wide range of proteins are secreted by highly specialized secretion systems. These secreted proteins play essential roles in the response of bacteria to their environment and also in several physiological processes such as adhesion, pathogenicity, adaptation and survival. Therefore, identifying secreted proteins in Gram-negative bacteria may assist in understanding the secretion mechanism and development of new antimicrobial strategies. Considering that a single-feature model is less likely to comprehensively cover this information, three kinds of feature models were used in this paper to represent protein samples by composition analysis, correlation analysis and smoothing encoding method on position-specific scoring matrix profiles. A support vector machine-based ensemble method with these hybrid features was developed to predict multi-type Gram-negative bacterial secreted proteins. Finally, our method achieves overall accuracies of 97.09% and 96.51% using an independent dataset test and jackknife test on a public test dataset, which are 3.49% and 2.32% higher, respectively, than results obtained by other methods. These results show the effectiveness and stability of the proposed ensemble method. It is anticipated that our method will provide useful information for further research on bacterial secreted proteins and secreted systems.
Background: Gram-negative bacteria interact with their environment by secreting a wide range of particular substrates (such as proteins) across two lipid bilayers from the cytoplasm to the extracellular space. Determining the types of secreted proteins is beneficial for further research on secreted proteins and secretion systems. Objective: As an essential alternative for experimental methods, an accurate machine learningbased multi-type Gram-negative bacterial secreted protein prediction method was proposed in this study. Methods: The main contribution is combining auto-cross-correlation analysis and feature ranking technology to build an effective support vector machine-based multi-type Gram-negative bacterial secreted protein predictor. The specifically designed auto-cross-correlation descriptor can capture evolutionary correlation information between amino acid pairs along protein sequence from position specific scoring matrices. Feature ranking technique was used to analyze and select the most informative features for building prediction model. Results: Several kinds of prediction accuracies obtained by independent dataset test are reported on two benchmark datasets. Compared with the state-of-the-art prediction methods, the proposed method improves overall accuracies by 2.91% and 2.25%, respectively. Conclusion: Our study will provide an important guide to utilize protein evolutionary information for further research on bacterial secreted proteins.
Background: Amino acid physicochemical properties encoded in protein primary structure play a crucial role in protein folding. However, it is not yet clear which of the properties are the most suitable for protein fold classification.Objective: To avoid exhaustively searching the total properties space, an amino acid properties selection method was proposed in this study to rapidly obtain a suitable properties combination for protein fold classification.Method: The proposed amino acid properties selection method was based on sequential floating forward selection strategy. Beginning with an empty set, variable number of features were added iteratively until achieving the iteration termination condition.Results: The experimental results indicate that the proposed method improved prediction accuracies by 0.26-5% on a widely used benchmark dataset with appropriately selected amino acid properties.Conclusion: The proposed properties selection method can be extended to other biomolecule property related classification problems in bioinformatics.