Removing reference bias and improving indel calling in ancient DNA data analysis by mapping to a sequence variation...

Area of Science:

• Genomics
• Bioinformatics
• Paleogenetics

Background:

• Ancient DNA (aDNA) analysis is crucial for studying past human populations.
• Degraded aDNA exhibits short sequences and post-mortem mutations, reducing mapping accuracy and causing reference bias.
• Reference bias favors mapping reads with reference alleles over non-reference ones.

Purpose of the Study:

• To evaluate the effectiveness of variation graph software (vg) in mitigating reference bias for aDNA analysis.
• To compare vg performance against existing methods like BWA for aDNA sequence alignment.

Main Methods:

• Alignment of simulated and real aDNA samples using vg against a variation graph.
• Comparison of vg alignments with BWA alignments to the human linear reference genome.
• Evaluation of variant detection sensitivity, especially for insertions and deletions (indels).

Main Results:

• Variation graph alignment with vg effectively removed reference bias, achieving balanced allelic representation.
• vg demonstrated more sensitive variant detection compared to BWA, particularly for indels.
• Alternative methods like relaxed BWA parameters or filtering reduced bias but were less sensitive than vg for indels.

Conclusions:

• Aligning aDNA sequences to variation graphs is an effective strategy to mitigate reference bias.
• Variation graph analysis retains mapping sensitivity and improves the detection of genetic variations, including previously missed indels.