IDRBP-PPCT: Identifying Nucleic Acid-Binding Proteins Based on Position-Specific Score Matrix and Position-Specific...

Area of Science:

• Molecular Biology
• Bioinformatics
• Computational Biology

Background:

• DNA-binding proteins (DBPs) and RNA-binding proteins (RBPs) are crucial nucleic acid-binding proteins (NABPs) involved in fundamental biological processes.
• Proteins binding both DNA and RNA (DRBPs) also play significant roles in gene expression regulation.
• Accurate identification of these proteins is vital for understanding protein-nucleic acid interactions.

Purpose of the Study:

• To develop an effective computational method for identifying DBPs, RBPs, and DRBPs solely from protein sequences.
• To introduce a novel protein representation technique incorporating evolutionary information and sequence correlations.
• To create a user-friendly web server for predicting DNA-, RNA-, and dual-binding proteins.

Main Methods:

• Proposed a novel protein representation method: Position-Specific Scoring Matrix (PSSM) and Position-Specific Frequency Matrix (PSFM) Cross Transformation (PPCT).
• Developed a computational predictor, IDRBP-PPCT, utilizing the PPCT method and a two-layer random forest algorithm.
• Integrated evolutionary information from PSSM and PSFM, along with their correlations, into the protein representation.

Main Results:

• The PPCT method effectively captures evolutionary and correlational information from protein sequences.
• The IDRBP-PPCT predictor demonstrated high accuracy in identifying DBPs, RBPs, and DRBPs on independent datasets.
• Validation on the tomato genome confirmed the method's effectiveness in a real-world biological context.

Conclusions:

• The PPCT representation method is effective for characterizing nucleic acid-binding proteins.
• IDRBP-PPCT provides a reliable computational tool for identifying DBPs, RBPs, and DRBPs.
• The developed web server offers accessible prediction capabilities for researchers studying protein-nucleic acid interactions.