CWLy-pred: A novel cell wall lytic enzyme identifier based on an improved MRMD feature selection method

Area of Science:

• Biochemistry
• Bioinformatics
• Enzymology

Background:

• Cell wall lytic enzymes are crucial in various research and industrial applications.
• Bioinformatic methods are increasingly used to streamline experimental processes and reduce costs.
• Accurate identification of these enzymes is essential for efficient research and development.

Purpose of the Study:

• To develop a novel machine learning-based tool for identifying cell wall lytic enzymes.
• To improve feature selection methods for enhanced model performance.
• To provide a user-friendly web application for accessing the developed tool.

Main Methods:

• Development of a support vector machine (SVM) based identifier named CWLy-pred.
• Application of an improved Minimum Redundancy Maximum Relevance (MRMD) feature selection method.
• Training and validation of the model using a curated dataset of enzyme features.

Main Results:

• CWLy-pred achieved high performance metrics: 93.067% accuracy, 85.3% sensitivity, 94.8% specificity, 0.775 MCC, and 0.900 AUC.
• The model demonstrated superior performance compared to existing state-of-the-art identifiers.
• The optimized feature set consists of only 6 dimensions, mitigating overfitting and enhancing experimental guidance.

Conclusions:

• CWLy-pred is a highly accurate and efficient tool for identifying cell wall lytic enzymes.
• The developed bioinformatics approach effectively reduces the need for extensive in vitro experimentation.
• The publicly available web application facilitates broader access and application in biological research.