Fully automated sequence alignment methods are comparable to, and much faster than, traditional methods in large data...

Area of Science:

• Bioinformatics
• Computational Biology
• Virology

Background:

• Multiple sequence alignment (MSA) is crucial for phylogenetic analysis but computationally intensive, especially with large and fragmentary genomic datasets.
• Traditional MSA methods involve automated alignment followed by manual curation, which is time-consuming and irreproducible at scale.
• Emerging fully automated MSA methods aim to address these limitations, but their compatibility with traditional approaches for phylogenetic inference is unclear.

Purpose of the Study:

• To compare the phylogenetic results obtained from traditional (automated + manual editing) and fully automated multiple sequence alignment methods.
• To assess the impact of different alignment strategies on hepatitis B virus (HBV) phylogenies, including analyses with whole genomes and sequence fragments.
• To evaluate the monophyly of existing HBV genotypes within phylogenies derived from various alignment approaches.

Main Methods:

• Utilized approximately 33,000 publicly available hepatitis B virus (HBV) sequences, including both whole genomes and fragments.
• Compared three MSA methods: purely automated traditional software, automated with manual editing, and recent fully automated approaches.
• Analyzed resulting tree topologies using multiple metrics and assessed the support for HBV genotype monophyly under different statistical thresholds.

Main Results:

• Traditional and fully automated alignment methods produced similar HBV phylogenies, with differences primarily attributed to phylogenetic uncertainty rather than alignment strategy.
• Fully automated alignment approaches were significantly less time-intensive and did not require manual intervention.
• Most diagnostic HBV genotypes did not consistently form evolutionarily distinct groups, irrespective of alignment type or support threshold.

Conclusions:

• Fully automated multiple sequence alignment algorithms are compatible with traditional methods and suitable for large, complex datasets, including fragmentary genomic data.
• Automated MSA significantly reduces the time and labor required for phylogenetic analysis.
• The findings suggest potential inaccuracies in current HBV genotype classification, necessitating a revision of the existing classification system.