DBETH: a Database of Bacterial Exotoxins for Human

Area of Science:

• Microbiology
• Bioinformatics
• Molecular Biology

Background:

• Pathogenic bacteria produce protein toxins to overcome host defenses.
• Advancements in genomics and structural biology enhance understanding of toxin mechanisms.
• Emerging bacterial species and genome rearrangements necessitate continuous toxin discovery.

Purpose of the Study:

• To systematically compile and analyze features of known bacterial toxins.
• To create a comprehensive knowledgebase for human pathogenic bacterial toxins.
• To develop a tool for identifying novel bacterial toxin sequences.

Main Methods:

• Developed the Database for Bacterial ExoToxins (DBETH).
• Integrated sequence, structure, and interaction network data for 229 toxins.
• Implemented a prediction server using homology and machine learning (support vector machines).

Main Results:

• DBETH contains data for 229 toxins from 26 bacterial genera, categorized into 24 types.
• The database provides sequence, structure, and physico-chemical property analyses.
• The prediction server identifies potential toxin sequences via homology and machine learning.

Conclusions:

• DBETH serves as a valuable resource for studying bacterial pathogenicity.
• The database facilitates systematic analysis and discovery of bacterial toxins.
• The integrated prediction server aids in identifying new toxin-like sequences.