BacSPaD: A Robust Bacterial Strains' Pathogenicity Resource Based on Integrated and Curated Genomic Metadata

Area of Science:

• Microbiology
• Bioinformatics
• Genomics

Background:

• Omics data in microbiology offers potential for studying bacterial pathogenesis.
• A comprehensive, curated resource cataloging bacterial strains and their human pathogenicity is lacking.
• Current methods for identifying pathogenicity determinants are often biased and incomplete.

Purpose of the Study:

• To address the gap in curated pathogenicity data.
• To introduce BacSPaD (Bacterial Strains' Pathogenicity Database), a resource for bacterial pathogenicity annotations.
• To facilitate the development of computational tools for predicting pathogenic potential.

Main Methods:

• Developed a rule-based annotation workflow.
• Combined trusted metadata, automated keyword matching, manual curation, and literature review.
• Curated a database of high-quality, complete bacterial genomes.

Main Results:

• Classified 5502 genomes as pathogenic to humans (HP) and 490 as non-pathogenic to humans (NHP).
• Included 532 species, 193 genera, and 96 families.
• Demonstrated a significant but moderate correlation between virulence factors and HP classification.

Conclusions:

• BacSPaD enhances understanding of bacterial pathogenicity mechanisms.
• The database aids in developing predictive models for bacterial pathogenic potential.
• A user-friendly web interface improves accessibility and provides visualization statistics.