Breakdancer Different Result

Question

[tahamasoodi]

Hi,

I'm using BreakDancer on the whole genomic data generated by Illumina Hiseq. In order to confirm that I'm running the program correctly, I tried to run it with the Yoruban genome (NA18507) which the developers have used in their algorithm. I downloaded the aligned bam file of NA18507 and used the following syntax:

./bam2cfg.pl -g -h /home/taha/Downloads/NA18507.mapped.ILLUMINA.bwa.YRI.low_coverage.20120522.bam > Examplenew_break.cfg
./breakdancer_max -c 4 -q 12 /home/taha/Softwares/breakdancer-1.1_2011_02_21/perl/Examplenew_break.cfg  > Examplenew_break_threash.xls

However after running, I got totally different results than what authors got. I got total 7200 variants while as they got 9202+4901+665 variants.

May I know why should be there such a difference?

Answer 1

[Chris Miller]

I imagine you're using a newer version than was used in the original paper. I believe that there have been many updates/improvements to the algorithm over the past few years.

Answer 2

[tahamasoodi]

But that does not mean that there may be difference of 8000 variants! How can I check the accuracy of BreakDancer results?

[Chris Miller]

1) You can use an orthogonal method (another algorithm)

2) For real validation of structural variants, we generally do capture, then deep sequencing of the breakpoint regions, then use an assembly-based process to confirm the presence of reads that support the breakpoint