A novel approach to estimating heterozygosity from low-coverage genome sequence

Area of Science:

• Population genetics
• Genomics
• Bioinformatics

Background:

• High-throughput sequencing enables population genetic parameter estimation.
• Heterozygosity is key for understanding inbreeding and population size.
• Low-coverage sequencing hinders accurate heterozygosity estimation due to genotype calling challenges.

Purpose of the Study:

• To develop a method for estimating genome-wide heterozygosity from low-coverage sequence data without intermediate genotype calling.
• To provide a robust method for population genetic analyses using widely available low-coverage data.

Main Methods:

• A novel method jointly learns allele distributions, sequencing error rates, and reference bias.
• The method bypasses the need for explicit genotype calling.
• Validation performed on simulated and real low-coverage sequence data.

Main Results:

• The method accurately estimates heterozygosity rates from low-coverage data, consistent with high-coverage estimates.
• Analysis of human populations revealed complex relationships between sequencing coverage and heterozygosity.
• Ratios of heterozygosity proved more interpretable and reliable than absolute estimates.

Conclusions:

• The developed method offers a reliable way to estimate heterozygosity from low-coverage sequencing data.
• Correcting for sequencing depth is essential for accurate heterozygosity estimation.
• Heterozygosity ratios are valuable for population genetic inferences, especially when comparing datasets.