filtering of read group

102 views
Skip to first unread message

Weihong Xu

unread,
Feb 11, 2013, 1:42:33 PM2/11/13
to igv-...@googlegroups.com
Hi,

I am using IGV 2.2.5 (7) and trying to mask some read groups in IGV. I tried to provide the flat text listing the group IDs in the "Preference", but no effect. Do you have any suggestion?

Say I have two read groups. Each line of bam record ends with either RG:Z:AAAA or RG:Z:BBBB. If I only want to show AAAA group, do I just make a file with BBBB?

thanks,
Weihong

Jim Robinson

unread,
Feb 14, 2013, 9:55:18 AM2/14/13
to igv-...@googlegroups.com
Hi Weihong,

What you are suggesting should work,  we will investigate.

Jim

Jim Robinson

unread,
Feb 14, 2013, 11:48:11 AM2/14/13
to igv-...@googlegroups.com
Could you send us your read group file, or a sample?


Roel Kluin

unread,
Sep 22, 2016, 6:50:24 AM9/22/16
to igv-help
Unless I'm mistaken, the filter based on read groups is still not working? Using IGV 2.3.81 (127), the current binary downloaded from the broad and jdk 1.8.0_101.


On Thursday, February 14, 2013 at 5:48:11 PM UTC+1, Jim Robinson wrote:
Could you send us your read group file, or a sample?


I have a file containing just one line IWFCDW, followed by a linux newline '\n'.

The filename was entered next to view->preferences->Alignments, with filter alignments by read group marked, you'll have to mark the check box a second time for some reason. doing so seems to trigger a refresh, but thereafter still all read groups are displayed (either when reads are grouped by read group or not, squished or expanded view).

The read group line in the bam file header is this format:

@RG    ID:IWFCDW    PL:ILLUMINA    LB:MERGED_4114_AAGGTACA_MANSC1_poolA.10_R1.fq.gz    SM:MERGED_4114_AAGGTACA_MANSC1_poolA.10_R1.fq.gz    CN:NKICMF

The bam files contains 96 different read groups in total. Reads have their respective RG:Z:IWFCDW tags, which are shown when hoovering over a read.


Thanks,
Roel


Jim Robinson

unread,
Sep 24, 2016, 11:37:49 PM9/24/16
to igv-help
Its working fine for me.   Try the attached filter file, and then load (by url) this bam: 

http://1000genomes.s3.amazonaws.com/phase3/data/HG01879/exome_alignment/HG01879.mapped.ILLUMINA.bwa.ACB.exome.20120522.bam

Jump to the gene "myc".   You should see nothing, because all reads are in the group ERR047763.    If you confirm this is working we will need to examine your bam file and read group filter file to see what's going on.   I can send instructions for extracting a small slice of your bam if needed.

BTW in future releases of IGV you will need to set the path to the read group file manually in the "prefs.properties" file, it is no longer included in the preferences GUI.  The entry should look like this

SAM.FILTER_URL=/Users/jrobinson/read_group_filter.txt
read_group_filter.txt
Reply all
Reply to author
Forward
0 new messages