CONGA: Copy number variation genotyping in ancient genomes and low-coverage sequencing data

Area of Science:

• Genomics
• Ancient DNA Analysis
• Evolutionary Biology

Background:

• Ancient genome studies traditionally focus on single nucleotide polymorphisms (SNPs).
• Copy number variants (CNVs) are crucial for disease and adaptation but challenging to detect in low-coverage ancient genomes due to short DNA fragments.
• Existing CNV detection software is ineffective with ancient DNA's typical low coverage (<1×) and short read lengths (<80 bps).

Purpose of the Study:

• To introduce CONGA, a novel computational tool specifically designed for genotyping CNVs in low-coverage ancient genomes.
• To demonstrate CONGA's accuracy and reliability in detecting deletions and distinguishing between heterozygous and homozygous states.

Main Methods:

• CONGA was developed and validated using simulations and down-sampling experiments on ancient genomes.
• The tool was applied to genotype 10,002 deletions across 71 ancient human genomes spanning 50,000 years.
• Analysis considered variable genome coverages (0.44×-26×) and read lengths (52-121 bps).

Main Results:

• CONGA can genotype deletions larger than 1 kilobase pair (kbps) with F-scores above 0.75 at ≥1× coverage.
• Technical factors like coverage and read length influenced deletion profiles in some samples (21/71).
• The majority of samples (50/71) showed coherent deletion frequencies, correlating strongly with SNP-based genetic diversity and exhibiting signatures of purifying selection.

Conclusions:

• CONGA provides a robust method for CNV analysis in ancient genomes, overcoming limitations of low and variable coverage.
• This facilitates systematic investigation of CNVs' roles in human evolution and adaptation.
• CONGA opens new avenues for understanding genetic diversity and selective pressures in ancient populations.