TAR-VIR

TAR-VIR reconstructs RNA viral strains from metagenomic sequencing data to enable strain-level characterization of viruses, including those lacking closely related reference genomes.

Key Features:

• Hybrid Approach: Integrates read classification with an in-house strain-level de novo assembly tool to reconstruct viral strains without requiring comprehensive reference genomes.
• Strain-Level Analysis: Performs high-resolution reconstruction and analysis of RNA virus strains to assess intra-species diversity and strain composition.
• Versatility: Provides a read recruiting stage that can operate standalone for strain reconstruction or supply recruited reads to other de novo assemblers for downstream functional and taxonomic analyses.
• Performance Validation: Validated on simulated and real viral metagenomic datasets and demonstrated competitive performance relative to existing tools.

Scientific Applications:

• Virology: Enables strain-resolved analysis of RNA viruses for studying viral diversity and pathogenicity.
• Epidemiology and Surveillance: Supports surveillance and outbreak investigation by identifying and reconstructing viral variants from metagenomic samples.
• Viral Evolution and Diversity: Facilitates reconstruction of intra-host and population-level viral diversity to inform evolutionary studies.
• Novel Virus Detection: Improves detection and reconstruction of viruses lacking closely related reference genomes, aiding discovery of novel or highly divergent strains.

Methodology:

Classify reads from metagenomic datasets using an effective read classification method, perform strain-level de novo assembly of the classified reads with an in-house assembler, and optionally use a read recruiting stage to provide reads to other de novo assembly tools.