Breakdancer Somatic Sv Input Order Of Tumor And Normal Bam Files

Question

[g.skullcap]

I am interested in using breakdancer to find somatic variants. The documentation shows an example of putting tumor and normal samples as inputs to breakdancer. Then does this mean that the output will contain somatic variants? We’ve tried inputting WGS tumor an normal samples in alternating order, i.e. Tumor-Normal and Normal-Tumor. The results are different, i.e. CTX counts are 227 for Tumor-Normal and 323 for Normal-Tumor. Does this mean that the order matters? How can I process/filter the outputs to get somatic variants? Thank you.

Answer 1

[Chris Miller]

According to this page, the input order should be tumor bam, then normal bam: http://breakdancer.sourceforge.net/pipeline.html

bam2cfg.pl -g -h tumor.bam normal.bam > BRC6.cfg

Via the author of breakdancer: No. The users will need to examine the origin of the supporting reads to determine if a variant is somatic.

Answer 2

[skm770]

Hi I used the breakdancer in the same order. Where the first sample was a treated condition with possibly multiple deletions and translocations but when I see the output it only shows supporting information from only the normal sample.

Here are the steps that I have tried to use.

My normal file is : AddOrRep_HN002.bam My Treated condition is : Sorted_nanomax.bam

I had used picard AddOrReplaceReadGroups.jar for both the files. Yet when I run the bam2cfg.pl using the order given in the manual for tumor and normal bam file. I get the following cfg file which does not seems to have information about the treated file but only has info about the normal sample :-

#Command : bam2cfg.pl -g -h treated.bam AddOrRep_HN002.bam > Sorted_nanomax.cfg

################################################## readgroup:HN002 platform:Illumina map:AddOrRep_HN002.bam readlen:90.00 lib:HN002 num:10001 lower:415.71 upper:535.14 mean:476.15 std:14.93 SWnormality:-21.23 flag:0(2.00%)1(0.14%)18(94.42%)2(1.0 5%)32(1.19%)4(1.05%)64(0.12%)8(0.04%)21142 exe:samtools view #######################################################

So when I run breakdancer-max I only get rows with readgroup information coming from the normal sample file. What should I do to get the that from the treated sample also. Here is the final output and command used.

#Command: breakdancer-max -q 10 -d Sorted_nanomax.ctx Sorted_nanomax.cfg

############################################ #Chr1 Pos1 Orientation1 Chr2 Pos2 Orientation2 Type Size Score num_Reads num_Reads_lib AddOrRep_HN002.bam Gm01 1 276+325- Gm01 2297 276+325- ITX -309 99 93 AddOrRep_HN002.bam|93 NA Gm01 1 135+184- Gm01 5794 124+121- ITX 261 99 36 AddOrRep_HN002.bam|36 0.34 Gm01 9717 123+97- Gm01 12876 123+97- ITX -255 54 7 AddOrRep_HN002.bam|7 NA Gm01 9717 110+84- Gm01 83483 209+273- ITX -268 99 55 AddOrRep_HN002.bam|55 1.45 Gm01 72653 33+1- Gm01 75695 29+33- DEL 2974 99 28 AddOrRep_HN002.bam|28 0.01 Gm01 76502 29+5- Gm01 78130 1+23- DEL 1526 99 17 AddOrRep_HN002.bam|17 0.09 Gm01 144470 33+34- Gm01 145494 33+34- DEL 177 99 31 AddOrRep_HN002.bam|31 NA #############################################

[Chris Miller]

I don't know the answer, but please open a new question with this info. TGI staff monitors biostar to provide support, but won't see your question buried here on an old post. Thanks.

[skm770]

I have opened it a a new question it can be found here : Problems with breakdancer