Classifying the Unclassified: A Phage Classification Method.

Area of Science:

• Microbiology
• Bioinformatics
• Genomics

Background:

• Accurate classification of bacteriophage genomes is crucial for understanding viral diversity and evolution.
• Existing methods may not fully capture the taxonomic diversity of phages, particularly unassigned genomes.
• Phages infecting Vibrionaceae represent a well-studied group, suitable for method validation.

Purpose of the Study:

• To introduce ClassiPhage, a novel computational method for classifying phage genomes.
• To validate ClassiPhage's performance using known phage families and Vibrionaceae-infecting phages.
• To assess the potential of ClassiPhage for classifying unassigned viral genomes and identifying novel phage lineages.

Main Methods:

• Development of ClassiPhage utilizing phage-specific Hidden Markov Models (HMMs) derived from protein clusters.
• Validation of ClassiPhage against publicly available phage genomes, including those infecting Vibrionaceae.
• Comparison of ClassiPhage classifications with existing taxonomic assignments from the International Committee on Taxonomy of Viruses (ICTV).

Main Results:

• ClassiPhage achieved classification consistent with ICTV assignments for well-described phage families (Myoviridae, Podoviridae, Siphoviridae, Inoviridae).
• 44 out of 58 unclassified Vibrio phage genomes were successfully assigned to a phage family.
• 14 genomes remained unassigned, suggesting potential novel phage families or subfamilies.

Conclusions:

• ClassiPhage provides a robust, genome-sequence-based approach for bacteriophage classification.
• The method demonstrates high accuracy for known phages and utility for classifying unassigned viral genomes.
• ClassiPhage has broad applications, including metagenomic analysis and prophage identification.