Assessing alignment-based taxonomic classification of ancient microbial DNA.

Area of Science:

• Palaeomicrobiology
• Metagenomics
• Bioinformatics

Background:

• Palaeomicrobiology is advancing rapidly with high-throughput DNA sequencing.
• Understanding ancient DNA characteristics is crucial for accurate microbial taxonomy in ancient samples.
• Current methods for assigning taxonomy to ancient metagenomic data need refinement.

Purpose of the Study:

• To investigate how ancient DNA characteristics influence alignment-based taxonomic classification.
• To assess the impact of DNA fragment length, damage, and divergence on classification accuracy.
• To evaluate the effect of reference database selection on taxonomic assignments in palaeomicrobiology.

Main Methods:

• Utilized simulated and published metagenomic datasets.
• Compared nucleotide-to-nucleotide and nucleotide-to-protein alignments for short DNA fragments (<60 bp).
• Assessed the impact of deamination and sequence divergence (∼100,000 years) on classification.
• Tested four different reference databases for taxonomic assignment.
• Reanalyzed published ancient dental calculus data.

Main Results:

• Nucleotide-to-nucleotide alignments are superior for short ancient DNA fragments.
• Deamination and significant genomic divergence had minimal impact on classification.
• Reference database choice significantly biased taxonomic classification outcomes.
• Reanalysis of dental calculus data increased taxonomic assignments by 64.2-fold and identified new species.

Conclusions:

• Ancient DNA characteristics significantly affect taxonomic classification accuracy.
• Nucleotide-to-nucleotide alignment is recommended for short ancient DNA fragments.
• Careful selection of reference databases is critical for reliable palaeomicrobial studies.
• This study provides key insights and recommendations for advancing palaeomicrobiology.