Evaluation of recombination detection methods for viral sequencing

Area of Science:

• Virology
• Computational Biology
• Evolutionary Genetics

Background:

• Viral recombination significantly influences viral evolution, shaping new populations and lineages.
• Accurate detection of recombination is essential for reliable evolutionary analyses, yet existing methods struggle with large datasets.
• The increasing volume of viral sequencing data necessitates scalable and accurate recombination detection methods.

Purpose of the Study:

• To assess the suitability of eight recombination detection methods (RDMs) for analyzing bulk viral sequencing data.
• To evaluate the performance and scalability of these RDMs using simulated and empirical viral sequencing data.
• To provide guidelines for validating recombination detection results and inform future method development.

Main Methods:

• Evaluated eight RDMs: PhiPack, 3SEQ, GENECONV, RDP, MaxChi, Chimaera, UCHIME, and gmos.
• Utilized simulated viral sequencing data with varying sequence diversities, recombination frequencies, and sample sizes.
• Analyzed empirical viral sequencing data to validate findings and demonstrate practical application.

Main Results:

• Assessed RDMs showed significant trade-offs between scalability, analytical approach, resolution, and accuracy.
• No single RDM was universally optimal; suitability depends on dataset properties (e.g., sequence diversity, recombination frequency).
• Performance varied considerably across methods when analyzing simulated and empirical large-scale viral sequencing data.

Conclusions:

• RDMs must be scalable and possess analytical capabilities appropriate for the research application.
• Method selection requires careful consideration of dataset characteristics to ensure accurate recombination detection.
• Guidelines for validation and insights into method limitations are provided for large-scale viral sequencing data analysis.