Software For Hla Typing Using Ngs?

Question

[mikhail.shugay]

Hello Everyone!

In our lab we're studying T-cell receptor repertoires and now are planning to complement this data with HLA-typing for huge cohorts of patients. I was wondering if there are people around here who do HLA typing using NGS can answer several questions.

1. Is there any reliable non-commercial and open software to determine HLA alleles from NGS data from Illumina platform?
2. Do you use commercial solutions, such as Omixon HLA typing?
3. What is your experience with commercial software that claims accurate processing of data from multiple platforms (Illumina, 454, IonTorrent, PacBio) like 157-P:NGSEngine. Is it really possible to correctly determine alleles with Illumina HiSeq reads?

Thanks in advance,

Mike

[Michael]

Just thinking this is very similar to whole exome sequencing: What is the best pipeline for human whole exome sequencing?, with the difference in the wet lab step to only enrich for exons of those HLA genes of interest. Could yield very high coverage of those regions, or give the ability to multiplex many samples (not sure if that is safe here!). The reliability should be comparable in accuracy or better than whole exome sequencing.

[mikhail.shugay]

I completely agree with you, we have also considered enriching for HLA loci. So the algorithms that are used to perform phasing for Exome-Seq data could also be used here. It could also be a good idea to incorporate HLA probes in enrichment kit for clinical Exome-seq in the future. E.g. a patient could be simultaneously screened for oncogenic SNPs while also being HLA-typed for a possible organ transplant.

Answer 1

[Sebastian Boegel]

If your are looking for a tool (i.e. algorithm) determining HLA types from standard NGS data (i.e. without any change to the seq lap protocol, e.g. targeted seq of MHC region etc), you might try those recently published tools applicable to all sorts of Whole xx Seq, such as whole genome, whole exome, RNA-seq, amplicon seq (Disclosure: i am author of one of these tools):

• HLAminer
• seq2HLA
• Athlates (already mentioned by Mikhail)
• HLAforest
• PHLAT

I hope this list helps.

Update (24/10/2014):

another method recently published in Bioinformatics called Optitype

[lh3]

Thanks for the update today!

[r0ntu]

Hi really quick question, I'm kinda new to this stuff. Do all of the above softwares (particularly HLAMiner and PHLAT) require RNASeq BAM files as input?

Answer 2

[Charles Warden]

This is a good question - high coverage alone won't be sufficient because (at least with Illumina data), you will have alignment issues due to high homology within the region.

This is a tool that a colleague of mine developed. I know it gave some results that looked decent, but I also think that particular PI ultimately decided to focus on non-HLA variants (so, maybe it still didn't match normal variant calls):

http://sourceforge.net/apps/mediawiki/hlaseq/index.php?title=Main_Page

Here are some other tool that I have heard of (but I haven't tried):

http://gatkforums.broadinstitute.org/discussion/65/hla-caller

http://www.biomedcentral.com/1471-2164/12/42 (paper for link above, I believe)

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

I remember some commerical option was considered at one time, but it turned out more than just NGS reads were needed (I think it needed to use special HLA amplicons and we had targeted sequencing data?). Not sure if that was Oximon.

I also heard that 454 was potentially better suited for HLA typing (due to longer reads), so the vendor might be able to provide soem software suggestions from that end.

[mikhail.shugay]

Thanks a lot for those links! GATK HLA Typer seems really worth a try, though it is no longer maintained anymore. I believe GATK HLA Typer could be used for enrichment-based protocols. neXtype paper and software also seems really interesting, although it requires to follow their custom protocol. As for 454 data, we are not quite fond of it, due to those indel errors. We are really looking forward to Illumina MiSeq 300+300 paired end which I believe would solve the problem of HLA allele phasing. By the way it would also allow to get full sequence of variable regions of antibody in a single read pair.

[GouthamAtla]

Did you try GATK HLA Caller ? Which Reference fasta file should we use ?

[iamxuyy]

Hello! Recently i am also working on the HLA typing for targeted sequencing data of Miseq 300+300 pair-end reads. Is there any suggestion? I did the HLA typing with the Hiseq 90+90 pair-end reads with SOAP-HLA and it can also work with the Miseq data.

http://soap.genomics.org.cn/SOAP-HLA.html

Silver

Answer 3

[mikhail.shugay]

Just adding a link to other quite promising software by Broad: http://www.broadinstitute.org/scientific-community/science/projects/viral-genomics/athlates

[gskbioinfo143]

Hi sir

at current scenario among Omixon, NGSengine, NxType which one is best for NGS HLA Typing using Ion Torrent Platforms

Thanks

Answer 4

[lh3]

A new HLA typing tool: http://biorxiv.org/content/early/2014/07/08/006973

[Sebastian Boegel]

Thanks for sharing. Interesting method. I was only wondering i) what is the input (they show an example which is WGS)? and ii) is the tool available?

[lh3]

The input is WGS. I don't know if the tool is available now. Just to make the list more complete. It is a bit unfortunate that all the available tools on WGS are not fully free.

Answer 5

[Alvaro Sebastian]

I just opened a Question asking for a training dataset for NGS HLA typing (reads >200bp from PCR amplicons).

Training dataset for NGS HLA typing (reads >200bp from PCR amplicons)

If someone is interested in the topic I've developed a software for it but I have only non model species data, so I cannot test it properly.

Answer 6

[Alejandro Jimenez Sanchez]

A new HLA typing software is HLAreporter. They claim to outperform PHLAT and HLAminer. It can be used with WES data and predicts both HLA class I and II.

It can be downloaded from their website but a password is required to uncompress the zip file.

[Ekaterina Esaulova]

Just in case - they provide password in manual.

Answer 7

[Folder40g]

Sorry for reviving back this topic.

My question is: Does anyone knows a data set with RNAseq for which the HLA typing has been performed by a precise method (PCR for instance)?

I would like to check a software but I've not been able to find in any paper (HLAmine, seq2HLA, optitype, etc) that the data the use also contains the HLA typed by a non-bioinformatic procedure.

Thanks

[Michael]

I think you should post this as a new question.

Answer 8

[Ömer An]

Here is a compiled list in a paper: https://www.nature.com/articles/jhg2015102/tables/2