Hla Typing Using Miseq Data

Question

[J.F.Jiang]

Hi all,

I am facing a problem of dealing with HLA typing using Miseq data. I want to get the HLA hyplotypes based on the Miseq data.

Using the target PCR method, we seq the target exon region which is less than the length of miseq reads, e.g., the exon=250bp, while the reads length is 300bp. In other words, the reads will cover all the exon nts.

For the diversity of the HLA hyplotypes, it is hard to directly get the results.

I have read some refs,

however, it seems that there is no direct soft to deal with such a thing. The HLACaller embeded in GATK is now unmatained by Broad institute, and the libraries have changed a lot due to the update of the HLA database.

Others suggested use the blast or mapping (bwa or bowtie) to get the results, I am trying to do such a thing, while I found that most reads can be mapped to many regions across the HLA genes, which make the results complicated to assembed into the hyplotypes.

Is there any good suggestions for this kind of data?

Thanks

Answer 1

[mikhail.shugay]

Hello! I've been checking this subject for a while. I suggest you to try out another tool from Broad Institute ATHLATES (http://www.broadinstitute.org/scientific-community/science/projects/viral-genomics/athlates) which works fine with exome data.

[J.F.Jiang]

Thanks, worth a try.

[J.F.Jiang]

it is a good tools after trying to typing the exome-seq data, however, if not all the exons are seqed, it will not be suitable. For example, using miseq to get only exon 2 & 3 for HLA class I genes.

Answer 2

[User 59]

The only specific software I know for this comes from Omixon:

http://www.omixon.com/hla/

As you realise the mapping is a non-trivial aspect for these loci.

[Charles Warden]

I agree - in fact, if memory serves me correctly, I think Omixon also required some specialized sample preparation. Either way, there as a relatively recent BMC Genomics protocol on HLA typing via MiSeq (but this requires a specific sample preparation). Both strategies are mentioned in this feed:

A: Software for HLA typing using NGS?

And here is the paper, for your convenience:

http://www.biomedcentral.com/1471-2164/15/63/abstract

[J.F.Jiang]

Both the tools had been checked before. As you mentioned, these tools are developed specifictly for their purpose, which needed some specialized sample preparation and format transformation. I think it can be set as ref, but may not good for me, since I am not good at code manipulating.

Thanks