A global characterization and identification of multifunctional enzymes

Area of Science:

• Biochemistry and Molecular Biology
• Bioinformatics
• Enzymology

Background:

• Multi-functional enzymes perform diverse physiological roles, essential for cellular communication and pathway coordination.
• Understanding these enzymes is critical for deciphering complex biological systems.
• Existing knowledge on multi-functional enzymes requires systematic characterization and novel identification methods.

Purpose of the Study:

• To systematically characterize known multi-functional enzymes structurally, functionally, and evolutionarily.
• To develop a predictive model for identifying novel multi-functional enzymes.
• To analyze the distribution and evolutionary patterns of multi-functional enzymes across different species.

Main Methods:

• Collected and analyzed 6,799 literature-reported multi-functional enzymes.
• Utilized physiochemical properties (charge, polarizability, hydrophobicity, solvent accessibility) for characterization.
• Developed a combined Support Vector Machine and Random Forest model for prediction.
• Analyzed enzyme distribution, evolutionary history, and associated metabolic pathways.

Main Results:

• Identified four key physiochemical properties crucial for multi-functional enzyme characterization.
• Successfully identified 6,956 potential novel multi-functional enzymes from the ENZYME database.
• Observed uneven distribution of multi-functional enzymes, with Bacteria possessing more than Archaebacteria and Eukaryota.
• Revealed gene gain and loss fluctuations during evolution and preservation in essential metabolic pathways.

Conclusions:

• Developed a robust computational approach for identifying novel multi-functional enzymes.
• Provided insights into the evolutionary dynamics and species-specific distribution of these enzymes.
• Established a publicly accessible database and prediction server for multi-functional enzymes.