Stability of SARS-CoV-2 phylogenies

Area of Science:

• Genomics
• Virology
• Bioinformatics

Background:

• The COVID-19 pandemic spurred rapid, real-time genetic sequencing of SARS-CoV-2, enabling evolutionary studies.
• Decentralized sequencing efforts across numerous labs introduce potential for systematic errors due to varied techniques.
• These errors can impact the accuracy of viral evolution and lineage tracing.

Purpose of the Study:

• To identify and characterize systematic errors in SARS-CoV-2 genome sequencing data.
• To assess the impact of these errors on phylogenetic inference and evolutionary studies.
• To develop tools and strategies for improving the quality and comparability of SARS-CoV-2 genomic data.

Main Methods:

• Analysis of recurrent mutations in SARS-CoV-2 genome sequences, focusing on lab-specific patterns and primer site co-localization.
• Phylogenetic inference to evaluate the effect of potentially erroneous mutations on viral lineage reconstruction.
• Development of computational tools for comparing, analyzing, and visualizing large SARS-CoV-2 phylogenies.

Main Results:

• Identified recurrent mutations predominantly observed by single labs, often near primer binding sites, potentially stemming from sequencing artifacts.
• Demonstrated that these artifactual mutations can significantly affect phylogenetic analyses, suggesting excess recurrent mutation or recombination.
• Developed and presented a toolkit for comparative phylogenetic analysis, error detection, and clade structure stabilization.

Conclusions:

• Systematic sequencing errors pose a significant challenge to accurate SARS-CoV-2 evolutionary inference and lineage tracing.
• The developed tools and proposed screening methods can help mitigate these errors, enhancing data reliability.
• Establishing a stable, shared clade structure is crucial for accurate scientific discourse and inference in SARS-CoV-2 evolution.