The polymorphic and repetitive nature of most of the HLA genes might bias the read mapping process, usually underestimating variability on highly polymorphic segments, or underestimating variability on others. Most mappers cannot properly handle HLA data, especially when using a single genome draft as a reference.
hla-mapper was designed to overcome these issues. It applies a scoring system based on known HLA sequences from the IPD-IMGT/HLA and published haplotypes, and locally curated sequences, sorting each pair of read according to their similarity from those sequences. Then, hla-mapper assigns each pair of read to the HLA gene it is most likely to be derived from.
The resulted gene-specific aligned files (BAM files) are more accurate than the ones generated by using directly the BWA or Bowtie2 aligners and a single reference genome.
hla-mapper is set to produce outputs for :
HLA-A, HLA-B, HLA-C, HLA-G, HLA-E, HLA-F
How to install:
The zip file contains precompiled binaries for MacOs and Linux 64bit.
Current hla-mapper version: 2.3
Current database version: 002.1
How to cite hla-mapper :
Hla-mapper: an application to optimize the mapping of hla sequences produced by massively parallel sequencing procedures. Human Immunology 2018. doi: 10.1016/j.humimm.2018.06.010