VirStrain

VirStrain identifies RNA virus strains from short-read sequencing data to provide strain-level resolution for genome epidemiology and transmission analysis.

Key Features:

• Strain-level analysis: Identifies single and multiple co-infecting strains from short-read sequencing data for strain-resolution studies.
• Greedy covering algorithm: Employs a greedy covering algorithm to derive unique k-mer combinations from highly similar reference genomes.
• Performance across diverse viruses: Validated on three types of RNA viruses with varying similarity distributions among their reference genomes.
• Benchmarking success: Demonstrates improved speed and accuracy over existing strain identification tools on simulated and real sequencing datasets.

Scientific Applications:

• Genome epidemiology: Identifies closest reference genomes to enable use of associated metadata, such as geographical locations, for inferring transmission networks.
• Infectious disease research: Detects multiple strains in a single host to support studies of co-infection dynamics, viral evolution, and implications for treatment.

Methodology:

Processes short-read sequencing data and uses a greedy covering algorithm to identify unique k-mer combinations that distinguish closely related viral reference genomes for strain-level identification, including detection of multiple strains.