I am testing out HLAscan for the first time and attempting to use it with WES NGS sequencing, and I am wondering if anyone has any advice for extracting HLA reads. I have the following method so far:
As the paper says, HLA specific reads should be extracted from exome reads before running the tool. As I have exome bams available (which were aligned using bwa mem) I extracted the reads as follows:
samtools view -b -hL ${bed_folder}/filtered.${gene}.bed $inbam > $outbam
filtered.*.bed was created by extracting all gene features related to my HLA genes of interest from the reference gtf, and converting them to a sorted and merged bed with bedtools.
==========
The reason I'm posting this question is that I read somewhere (can't find the paper) that due to the high sequence diversity of the HLA region, mapping to the HLA genes can be difficult. I am worried that HLA mapping reads would either go unmapped or go to one of the decoy regions (my bams are aligned using hs37d5, which includes decoy sequences). Does anyone have experience extracting HLA reads from NGS sequencing that they could share?
You can also compare the results using Optitype. I have used in the past, pretty straightforward. https://github.com/FRED-2/OptiType
i'm looking for a genotyper for class ii alleles as well. unfortunately, optitype does not support that.