Unmapping Alternative reads

Question

[godth13teen]

Hi, I want to do some HLA typing, most of the tools require the bam file is aligned to primary genome without alternative read handling. From the International Genome Sample Resource project, I can get the high coverage file, but they are all aligned to primary genome with hla, decoy and alternative reads.

I want to unmap those alternative reads, I tried this script https://github.com/humanlongevity/HLA/blob/master/bin/get-reads-alt-unmap.sh, but there are some technical problems, so I want to know is there any other way or tools to do this task?

Thank you very much.

[GenoMax]

there are some technical problems

What kind of problems. It may be easy to solve them.

[godth13teen]

Hi,

I made a github issue in here: https://github.com/humanlongevity/HLA/issues/51

In short, there's an [E::bwa_idx_load_from_disk] fail to locate the index files error, and the script output nothing

[GenoMax]

Have you created an index for the BAM files you downloaded (samtools index, you would likely need to samtools sort before that)? If you use latest samtools then sort --write-index to do this in one step.

[godth13teen]

yes, I made the index *.bam.bai for the bam file beforehand

[GenoMax]

Looks like you did not name sort (samtools sort -n) your files before indexing them.

[godth13teen]

The bam file is already sorted:

samtools view -H HG02082.final.bam |grep "@HD"
@HD     VN:1.5  GO:none SO:coordinate

[godth13teen]

but you mean that I need to sort it by name, not coordinate, am I right?

[GenoMax]

Based on the errors you posted in GitHub issue, yes.

[godth13teen]

I tried sorting it by queryname and index it, but the script still does not work