"ERROR: No read groups found" when running allelotype

[Yen Lung Lin]

Hi,

I am trying to run the allelotype lobSTR_4.0.6 on a hg38 bam file generated with BWA_MEM. 

I used the command suggested in the documentation:

allelotype \
   --command classify \
   --bam my_sample_output.sorted.bam \
   --index-prefix hg19_v3.0.2/lobstr_v3.0.2_hg19_ref/lobSTR_ \
   --strinfo hg19_v3.0.2/lobstr_v3.0.2_hg19_strinfo.tab \
   --noise_model models/illumina_v3.pcrfree \
   --out my_sample_output

The bam file was aligned earlier using BWA-MEM. The index and strinfo were built by me using data downloaded from the Tandem Repeat Database. The noise_model was the same as the example.

I received the error message "ERROR: No read groups found". 

I also try to add the --realign flag but nothing change. I am wondering if someone knows what happens? Thanks in advance.

[Yen Lung Lin]

I am here to provide more information regarding this error:

I can see the @RG from the header of my input bam file. I am wondering if the error message has anything to do with the bam file itself or not?

[Melissa Gymrek]

Hi Yen,

First, lobSTR has not been maintained in a while, so I highly recommend you use HipSTR instead: https://github.com/tfwillems/HipSTR.

For this error, it's possible to have a @RG tag in the header but the individual reads might not be annotated with read groups. In any case I think it best to use HipSTR.

Best,

~Melissa

--
You received this message because you are subscribed to the Google Groups "lobSTR user group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to lobstr-user-gr...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/lobstr-user-group/e054e8f5-b65c-4e98-a39d-7d69c1e3c29b%40googlegroups.com.

[Yen-Lung "Onta" Lin]

Thank you for your answer. I have been realigning the BAM file to ensure the RG flag meets the requirement. It works though it's computational expensive. I will go try HipSTR as you recommended. 

Best,

Onta

To view this discussion on the web visit https://groups.google.com/d/msgid/lobstr-user-group/CAOvuif%2BSp%2BuNvJkDiprY8Xejwd0sKWM61gca0owtCV4GHLEjfg%40mail.gmail.com.

[Brendan Kohrn]

If all the reads are the same readgroup, you could use AddOrReplaceReadgroups from Picard tools ( https://broadinstitute.github.io/picard/) or GATK ( https://software.broadinstitute.org/gatk/documentation/tooldocs/4.0.0.0/picard_sam_AddOrReplaceReadGroups.php); that might be less computationally expensive.  

To view this discussion on the web visit https://groups.google.com/d/msgid/lobstr-user-group/CAMJis5F4s7E9%2Bg0S2DLD45QGvYncLgaWdZXCG7F0vQEyiJ9Baw%40mail.gmail.com.