How to map by bwa with 50 mismatches?

Question

[agata88]

Hi all,

I am trying to map amplicon PR reads (2x150bp) to gene reference by bwa. I would like to allow bwa align reads with 50 mismatches in first 150bp read and the same for second read.

Could you please help me find an option or algorithm that will be able to do that?

Thanks in advance,

Best,

Agata

[WouterDeCoster]

I wonder where those mismatches come from, and I believe you are trying to achieve something that probably has a better solution. What is the aim of this project? What is sequenced?

[GenoMax]

If this is amplicon seq why do you need to allow 1/3rd mismatches as compared to the length of the reads?

[H.Hasani]

Are you trying to induce artificial mutations to these regions?

[agata88]

This is for HLA typing. So I have very polymorphic regions of different genes in one sample. I don't know yet if this is a good solution, I would like to test that.

[GenoMax]

In that case take a look at: https://omictools.com/hla-typing-category
Software For Hla Typing Using Ngs?

[agata88]

Which one of the program listed here would you recommend for Amplicon-seq?

[agata88]

Ohh, sorry I deleted posts...can you anyhow bring them back?

[agata88]

Thanks, but I am not looking for a ready to go program :) Do you know which option of bwa can I use?

[WouterDeCoster]

Please use ADD COMMENT/ADD REPLY to answer to earlier answers or comments to keep this thread logically structured and easy to follow.

[GenoMax]

Sounds like you don't want to use "right tool for the job" :-)

If you want to use bwa then go for it. You could start with smaller seeds (-k) than default, drop penalty for mismatches (-B) etc. Could try the ready made -x settings for (PacBio/Ont2d) as a start.

[agata88]

I am little confused... I need to analyse KIR and HLA genes ...most of the programs analyse only few HLA genes, and diving into whole pipelines is more time consuming that writing own pipeline (since the procedure is not much complicated). So, I wanted to repeat this methodology:

https://genomemedicine.biomedcentral.com/articles/10.1186/gm396

and use it for all genes that I need in MY analysis.

[Chris Cole]

Yet in that paper they don't use BWA either. Why not use their methodology and indeed their available software rather than re-invent the wheel?

[agata88]

Ah, sorry, not this paper:

here is the proper one:

http://bmcgenomics.biomedcentral.com/articles/10.1186/1471-2164-14-355

[GenoMax]

Another example of the incomplete information that accompanies bioinformatics sections in papers. You could use bwa with just default parameters to see what you get. Perhaps the authors just used defaults and that is the reason no special mention has been made of detailed settings.