Setting @RG ID, PL, PU, LB, SM etc.

[alexbmp]

Hi all

When running bwa, I want the read group information to be implanted so that I can use the information in picard and GATK later.

I wrote down what I understood. Please point out if you think I'm wrong about something

-I see that RG ID is a read discriminator; even when BAM files are merged thanks to RG ID the reads can be discriminated.

-RG PL seems to be the sequencing machine platform (e.g. illumina).

-RG LB seems to be referred by picard MarkDuplicates so PCR duplicates in each sequencing library can be removed.


Now, I mention here that I read SAM-format spec.s.
( http://samtools.sourceforge.net/SAM1.pdf )
However, I'm confused about 2 more things.

1. Why does RG PU exist in the first place? Currently, the only reason I put in PU is to avoid picard and GATK errors later.

2. What's the difference between RG ID and RG SM? Is there any sutble difference between the two?

Thanks for your replies in advance!


Have a great day

[maubp]

SM is sample, and one sample might be sequenced twice, for instance with 454 and Illumina, which would be two read groups with different platform. At least, that is my understanding.

[alexbmp]

Thank you for your reply. So although I don't need the information right now, it can be important in some other pipeline I might take, right?

SM is sample, and one sample might be sequenced twice, for instance with 454 and Illumina, which would be two read groups with different platform. At least, that is my understanding.