read group: GATK or BWA option?

[m_elena_bioinfo]

Hi NGS users,
when I run GATK recalibration score, it returns me the error about read group:

SAM/BAM file SAMFileReader{/dati1/Analysis_NGS/prova_s8/prova_GATK_localrealignment.bam.sorted.bam} is malformed: The input .bam file contains reads with no read group.

So I try to use the option with a random string and the program runs:

--default_platform Illumina --default_read_group HWUSI-EAS703

I don't know if this is correct or if I have to rerun bwa, using the patch and use this options:

-i read group identifier (ID)
-m read group sample (SM), required if ID is given
-l read group library (LB)
-p read group platform (PL)

Any suggestions?
Thanx a lot,
ME

[aleidenroth]

Hi,

I had this issue, and after patching bwa to the lastest update I run my sampe with

-r STR read group header line such as `@RG\tID:foo\tSM:bar' [null]

I specify ID, SM, LB and PL. So far no trouble with GATK on the downstream files.

[mpiro]

It depends!

If you have one lane/group, using the GATK default_read_group is as good as using the patched bwa.
But if you are analyzing multiple lanes/groups, patched bwa should be used. The GATK recalibration requires different group name in order to analyze each group seperately and default_read_group only assign one default group name to all reads. This can affect the GATK recalibration score.

.

[Aicen]

malformated @RG line

Hi NGSer,
i run into the problem that my Readgroup Header is not accepted by bwa and i dont know why

heres my code:

bwa sampe -r @RG ID:ILLUMINA-52179E_0039_FC62HDBAAXX_1_1 SM:48_2 PL:illunmina ./testingscripts/chrY.fa ./testingscripts/48_2_1_KESC1_mymod.sai./testingscripts/48_2_2_KESC1_mymod.sai ./testingscripts/48_2_1_KESC1_mymod.fastq ./testingscripts/48_2_2_KESC1_mymod.fastq > ./testingscripts/chrYvs48_2_1_KESC1_mymod_48_2_2_KESC1_mymod.sam

any help would be nice . I also tested with \t explicit between the tags but with no success , thx in advance Aicen

[ulz_peter]

you should separate the tags by tabulators instead of whitespaces just put \t between the tags
like:

bwa sampe -r "@RG\tID:ILLUMINA-52179E_0039_FC62HDBAAXX_1_1\tSM:48_2\tPL:illumina" ./testingscripts/chrY.fa ./testingscripts/48_2_1_KESC1_mymod.sai./testingscripts/48_2_2_KESC1_mymod.sai ./testingscripts/48_2_1_KESC1_mymod.fastq ./testingscripts/48_2_2_KESC1_mymod.fastq > ./testingscripts/chrYvs48_2_1_KESC1_mymod_48_2_2_KESC1_mymod.sam

[Aicen]

" - this missed

my string missed the " at start and end and i needed to write \\t in my script so that the command lane statement went right, thx for your help

[rdeborja]

With BWA output, I usually use Picard's AddOrRemoveReadGroups.jar before feeding the BAM file to GATK. You can provide all the necessary read group details to ensure GATK works.

[Aicen]

bwa

Well, at the moment it seems to work fine, did u make other experience where bwa's -r opton could fail ? Because i dont want to draw the short stick at the end of the work realising sth. didnt work with the RG header^^ .

Iam using GATk for further variant calling btw

[alex_dl]

To have a bam that can be used in GATK, use this RG header (note the "\t")

bwa sampe -r @RG"\t"ID:SAMPLE1_RG1"\t"PL:illumina"\t"PU:SAMPLE1_RG1_UNIT1"\t"LB:SAMPLE1_LIB1"\t"SM:SAMPLE1

[prasadg]

you should separate the tags by tabulators instead of whitespaces just put \t between the tags
like:

bwa sampe -r "@RG\tID:ILLUMINA-52179E_0039_FC62HDBAAXX_1_1\tSM:48_2\tPL:illumina" ./testingscripts/chrY.fa ./testingscripts/48_2_1_KESC1_mymod.sai./testingscripts/48_2_2_KESC1_mymod.sai ./testingscripts/48_2_1_KESC1_mymod.fastq ./testingscripts/48_2_2_KESC1_mymod.fastq > ./testingscripts/chrYvs48_2_1_KESC1_mymod_48_2_2_KESC1_mymod.sam

Hello ulz_peter,

I tried writing code according to you , But didnt get success.

@HWI-ST741:204: D0TEJACXX:8:1101:1107:1901 1:N:0: BC=ACGTAA SAMPLE=1 LENGTH=64 MEAN_QUAL=38.6

Here is my sample Can you please help me writing the code for this sample.

Thanking in anticipation.