Mitochondrial DNA Consensus Calling and Quality Filtering for Constructing Ancient Human Mitogenomes: Comparison of...

Area of Science:

• Ancient Genomics
• Paleogenetics
• Bioinformatics

Background:

• Ancient DNA (aDNA) retrieval faces challenges like low molecule counts, fragmentation, and contamination, complicating sequence reconstruction.
• High-throughput sequencing platforms require robust methods for accurate consensus aDNA sequence generation.

Purpose of the Study:

• To evaluate and compare the performance of ANGSD and schmutzi for reconstructing high-quality ancient human mitochondrial genomes (mitogenomes).
• To assess the impact of quality filters and sequence parameters on ancient mitogenome reconstruction accuracy.

Main Methods:

• Retrieved raw genomic data (BAM/FASTQ) from 17 ancient human genomes (14th to 7th millennium BCE).
• Reconstructed mitochondrial consensus sequences using ANGSD and schmutzi with varying quality filters.
• Analyzed sequence parameters (read count, coverage, contamination) as predictors of sequence quality.

Main Results:

• Complete mitogenomes were successfully reconstructed for all samples; filtering improved accuracy for most.
• Schmutzi outperformed ANGSD in moderate to high coverage samples (>1,000,000 reads) by accounting for contamination.
• ANGSD showed better performance on low-quality samples due to its read termini trimming filter.
• Number of sequence reads and bases were the strongest predictors of overall sample quality.
• A novel haplogroup (U3b) was identified in an Early Chalcolithic individual.

Conclusions:

• Both ANGSD and schmutzi are effective for ancient mitogenome reconstruction, with distinct strengths for different data qualities.
• Quality filtering significantly enhances ancient mtDNA consensus calling and haplogroup prediction.
• The study identifies key predictors of ancient DNA sequence quality and reports a new haplogroup.