Efficient and accurate KIR and HLA genotyping with massively parallel sequencing data.

Area of Science:

• Immunogenetics
• Computational Biology
• Genomics

Background:

• Killer cell immunoglobulin like receptor (KIR) and human leukocyte antigen (HLA) genes are crucial for immune responses.
• These genes are highly polymorphic and challenging to genotype accurately using standard methods.
• Existing tools often fail to resolve complex KIR gene structures and variations.

Purpose of the Study:

• To develop a novel computational method, T1K, for efficient and accurate KIR and HLA allele inference.
• To address the limitations of current genotyping tools, particularly for highly similar and absent KIR genes.
• To enable comprehensive immunogenetic analysis across diverse sequencing data types.

Main Methods:

• T1K utilizes a novel computational approach for inferring KIR and HLA alleles.
• The method jointly analyzes alleles across all genotyped genes.
• It is applicable to RNA-seq, whole-genome sequencing (WGS), and whole-exome sequencing (WES) data.

Main Results:

• T1K reliably identifies gene presence and distinguishes homologous genes, including difficult KIR loci like KIR2DL5A/B.
• The method achieves high accuracy in HLA genotyping benchmarks.
• T1K successfully calls novel single-nucleotide variants and processes single-cell data.

Conclusions:

• T1K offers an efficient and accurate solution for KIR and HLA genotyping.
• The method's ability to handle complex gene structures and diverse data types expands genotyping capabilities.
• Application to single-cell RNA-seq revealed enriched KIR2DL4 expression in tumor-specific CD8+ T cells, highlighting T1K's potential in translational research.