SNBRFinder: A Sequence-Based Hybrid Algorithm for Enhanced Prediction of Nucleic Acid-Binding Residues

Area of Science:

• Biochemistry and Molecular Biology
• Bioinformatics and Computational Biology
• Genomics and Proteomics

Background:

• Protein-nucleic acid interactions are crucial for fundamental biological processes.
• Accurate identification of DNA- and RNA-binding residues in protein sequences is increasingly important.
• Current feature-based prediction algorithms require improved accuracy.

Purpose of the Study:

• To develop a novel sequence-based hybrid algorithm for predicting nucleic acid-binding residues.
• To enhance the accuracy and reliability of DNA- and RNA-binding residue identification.
• To provide an accessible web server for the SNBRFinder tool.

Main Methods:

• Developed SNBRFinder, a hybrid algorithm merging a feature predictor (SNBRFinderF) and a template predictor (SNBRFinderT).
• SNBRFinderF utilizes support vector machines with sequence profiles and descriptors.
• SNBRFinderT employs profile hidden Markov models for template detection in sequence alignment.

Main Results:

• SNBRFinderF outperformed commonly used sequence profile-based predictors.
• SNBRFinderT demonstrated performance comparable to structure-based template methods.
• The hybrid SNBRFinder significantly improved DNA- and RNA-binding residue predictions, achieving competitive performance with structure-based methods.

Conclusions:

• SNBRFinder offers significant advantages over existing sequence-based prediction algorithms.
• The sequence-based hybrid approach provides accurate and reliable prediction of nucleic acid-binding residues.
• An easy-to-use web server is available at http://ibi.hzau.edu.cn/SNBRFinder for broader accessibility.