A SNP panel for co-analysis of capture and shotgun ancient DNA data

Area of Science:

• Genetics
• Bioinformatics
• Computational Biology

Background:

• Technological advancements have led to an exponential increase in ancient human genome data generation.
• Platform-specific biases in sequencing technologies hinder the co-analysis of ancient and modern human DNA data.
• Integrating data from diverse sequencing platforms and ancient individuals presents significant analytical challenges.

Purpose of the Study:

• To develop a method for identifying single nucleotide polymorphisms (SNPs) with minimal technology-specific bias for ancient DNA analysis.
• To create a standardized SNP panel for improved co-analysis of ancient and modern human genomic data.
• To facilitate joint analyses across different sequencing technologies and populations.

Main Methods:

• Leveraged whole genome data from over 16,300 ancient individuals to identify SNPs with low technology-specific bias.
• Developed the "Compatibility" panel of one million SNPs suitable for ancient DNA experiments.
• Identified the "Compatibility-HO" panel, a subset of SNPs also genotyped on the Affymetrix Human Origins array for modern human data integration.

Main Results:

• The "Compatibility" panel significantly reduces spurious Z-scores caused by different sequencing platforms by nearly an order of magnitude.
• The panel retains substantial statistical power (65-85%) for f-statistic analysis.
• The identified SNP panels are broadly applicable to a large fraction of published ancient human DNA studies.

Conclusions:

• The "Compatibility" panel effectively mitigates technology-specific biases in ancient DNA data analysis.
• This standardized SNP set enhances the co-analyzability of ancient and modern human genomic datasets.
• A user-friendly tool is provided to balance bias reduction and statistical power for specific research needs.