DeEPn: a deep neural network based tool for enzyme functional annotation.

Area of Science:

• Biochemistry
• Bioinformatics
• Computational Biology

Background:

• High-throughput techniques accelerate enzyme sequence discovery.
• Accurate functional annotation of enzymes is crucial for understanding biological roles.
• Existing enzyme functional class prediction models often fail to address the seventh functional class (EC7).

Purpose of the Study:

• To develop a precise computational model for classifying enzymes across all seven functional classes.
• To address the limitations of current models in predicting EC7 enzymes.
• To introduce DeEPn, a Deep Neural Network-based approach for enzyme functional classification.

Main Methods:

• Development of a Deep Neural Network (DNN) model named DeEPn.
• Training and validation of DeEPn using enzyme sequence data.
• Comparative analysis of DeEPn against established tools like ECPred and SVM-Prot.

Main Results:

• DeEPn demonstrates high precision and accuracy in classifying enzymes across all seven functional classes (EC1-EC7).
• DeEPn significantly outperforms ECPred and SVM-Prot in predictive quality.
• The DeEPn tool is available as a web server for public use.

Conclusions:

• DeEPn provides a robust solution for accurate enzyme functional classification, including the challenging EC7 class.
• The developed model enhances the ability to decipher biological roles of newly discovered enzymes.
• DeEPn represents a significant advancement in bioinformatics tools for enzyme annotation.