Mapping paired end and single end reads together

Question

[GR]

Hi Everyone,

I have my raw data as two paired end reads file (one for left and one for right reads) plus a file where I have only single end reads. I am wondering if it is possible to align all this data to the reference genome?

Thanks.

[iraun]

Yes, it is possible. Which kind of data do you have? RNASeq or DNA?

[Devon Ryan]

Whether this is possible in a single command is completely dependent on the aligner you're using.

[GR]

I have DNA seq data and I want to use bwa.

[samuel]

I was thinking of doing the same thing but won't we be duplicating reads, as we will be counting the same read twice - once from paired end fastq and once from the single read fastq???

Answer 1

[Gabriel R.]

Is this genomic? Is BWA ok? You could transform the data into BAM and use this: https://github.com/udo-stenzel/network-aware-bwa

It's a custom version of BWA that eats BAM (paired and single reads) and spits out BAM.

[GR]

I am wondering if it would be appropriate to do the following: 1) map the paired end and single end read data individually 2) sort the data 3) Merge both the bam files (single end+paired end).

Please suggest!

[Devon Ryan]

That's perfectly fine.

[Gabriel R.]

I would do this instead:

1. Merge both the paired and unpaired using samtools cat
2. Pipe into network-aware-bwa

This gives you a single mapping step and does not require additional files.

[GR]

Thanks Samuel. I will try this.

[Gabriel R.]

You need libzeromq to install the bwa package. You can get it via apt-get install