Locityper starts with a collection of known locus haplotypes (alleles), and a given WGS dataset. It then evaluates all haplotype pairs (genotypes), and searches for the most likely one. Optimal genotype is selected based on three factors:

• Minimal number of sequencing errors,
• Optimal insert sizes (for paired-end data),
• Optimal read depth profile: no dips and excesses of read coverage.

You can find more details in our preprint (opens in a new tab).