Best practices for improving alignment and variant calling on human sex chromosomes

Area of Science:

• Genomics
• Human Genetics
• Bioinformatics

Background:

• Sex chromosomes (X and Y) share similarities with autosomes but have unique characteristics.
• Standard genomic analyses may not adequately process sex chromosome variations.
• Pseudoautosomal regions (PARs) and X-transposed regions (XTR) present specific challenges.

Purpose of the Study:

• To evaluate the impact of sex chromosome complement-informed strategies on variant detection.
• To compare standard autosomal analysis with specialized sex chromosome analysis.
• To identify best practices for accurate variant calling on sex chromosomes.

Main Methods:

• Simulation study assessing alignment, variant calling, and filtering strategies.
• Comparison of autosomal versus sex chromosome complement-informed reference genomes.
• Analysis of variant calling accuracy (true and false positives/negatives) in XX and XY samples.

Main Results:

• Sex chromosome complement-informed alignment increases true positives in PARs and XTR (XX samples).
• Masking XTR in XY samples leads to a ten-fold increase in false positives.
• Haploid calling in XY samples reduces false positives without increasing false negatives.

Conclusions:

• Accurate variant calling on sex chromosomes can reveal disease-relevant genetic variants.
• Recommended best practices include using sex chromosome complement-informed reference genomes and accurate ploidy parameters.
• Improved sex chromosome variant detection aids studies in health and disease.