PVGA: a precise viral genome assembler using an iterative alignment graph.

Area of Science:

• Genomics
• Bioinformatics
• Virology

Background:

• Viral genome analysis is critical for understanding virus evolution and mutation, especially since the COVID-19 outbreak.
• RNA viruses exhibit high mutation rates, where single-nucleotide variations can significantly alter viral function and pathogenicity.
• Specialized assembly methods are essential for accurate viral genome analysis due to these inherent complexities.

Purpose of the Study:

• To develop a novel viral genome assembler that integrates assembly and polishing into a unified workflow.
• To enhance the accuracy of viral genome reconstruction, enabling the detection of subtle genomic variations.
• To provide a reliable solution for advancing the understanding of viral evolution and behavior.

Main Methods:

• PVGA utilizes a reference genome and sequencing reads to construct an alignment graph.
• Dynamic programming determines the optimal genomic path by maximizing read support density.
• The process iteratively refines the genome using updated references until no further improvement is achieved.

Main Results:

• PVGA consistently outperforms existing programs in viral genome assembly quality across simulated and real datasets (long and short reads).
• The method demonstrates high accuracy, correctly identifying true genomes with zero mismatches and indels on simulated Nanopore data.
• PVGA's running time is comparable to other popular viral genome analysis tools.

Conclusions:

• PVGA offers a novel, integrated approach to viral genome assembly and polishing.
• The tool's high accuracy facilitates the detection of critical genomic variations impacting viral function and pathogenicity.
• PVGA provides a precise solution for advancing research into viral evolution and behavior.