PhaGCN2

PhaGCN2 classifies viral genomic sequences at the family level to enable accurate taxonomic assignments of novel and fragmented viral genomes.

Key Features:

• Rapid Classification: Performs rapid family-level classification of viral sequences.
• Enhanced Precision and Recall: Achieves higher precision and recall compared with vConTACT2, CAT, and VPF-Class, validated using standard evaluation metrics.
• Scalability: Increases the number of classifiable viral sequences, for example a fourfold expansion for the Global Ocean Virome dataset (v2.0) and classification of over 90% of the Gut Phage Database.
• Database Expansion Support: Enables high-throughput automatic classification to assist expansion of the International Committee on Taxonomy of Viruses (ICTV) database.
• Visualization Capabilities: Provides visualization of family-level associations among viruses to elucidate viral relationships.

Scientific Applications:

• Virology research: Supports accurate classification of novel viruses and fragmented genomes in virology studies.
• Large-scale virome analysis: Applies to large-scale datasets such as the Global Ocean Virome (v2.0) and the Gut Phage Database for taxonomic assignment.
• Ecology and evolution: Facilitates ecological studies and investigations of viral diversity and evolution by providing family-level taxonomic labels.

Methodology:

Employs advanced algorithms for viral sequence classification and is evaluated against established benchmarks.