Computational Identification and Analysis of Ubiquinone-Binding Proteins

Area of Science:

• Biochemistry and Molecular Biology
• Bioinformatics
• Computational Biology

Background:

• Ubiquinone is a crucial cofactor in numerous biological processes.
• Ubiquinone-binding proteins (UBPs) are essential for ubiquinone function.
• Computational identification of UBPs can illuminate ubiquinone-related pathways.

Purpose of the Study:

• To develop and validate UBPs-Pred, a novel computational predictor for identifying UBPs.
• To enhance the understanding of ubiquinone-binding proteins and their biological roles.
• To leverage machine learning for predicting protein function.

Main Methods:

• Feature selection from sequence-derived attributes.
• Development of the UBPs-Pred tool using the XGBoost classifier.
• Optimization of XGBoost parameters via Multi-Objective Particle Swarm Optimization (MOPSO).
• Bioinformatic analysis including motif statistics, protein distribution, Gene Ontology (GO), and KEGG pathway enrichment.

Main Results:

• UBPs-Pred demonstrated significant prediction performance with a Matthews correlation coefficient (MCC) of 0.517 on independent validation.
• Identified key sequence-derived features for accurate UBP prediction.
• Bioinformatic analyses provided insights into UBP binding domains, distribution, and associated biological pathways.

Conclusions:

• UBPs-Pred offers a reliable computational method for identifying ubiquinone-binding proteins.
• The study provides valuable insights into the functional roles and pathways of UBPs.
• This work contributes to the field of bioinformatics and computational biology by offering a new tool for protein function prediction.