Breakdancer Options For Minimum Quality Mapping

Question

[madkitty]

I'm trying to find CNV of our whole genome samples (split into 4 bam files). The first time I ran BAM2CFG and Breakdancer with NO option at all, and the output gave me interesting somatic CNVs also a huge collection of long CNVs and repeated CNVs which doesn't look too good. So I decided to trim out a bit and keep only good mapping quality CNVs. So far I was selecting with sed whatever has a mapping quality of 60 into each bam files. But then the CNV output was shorten to only 10% of what I had previously .. So doing again a selection with SED of reads mapped with a lower quality is gonna be a pain.

Can I use Breakdancer -m option to focus only on reads with a minimum mapping quality ? If I write the following :

breakdancer_max -m 40 file1.cfg > cnv.txt

Is it gonna for CNV among reads having a mapping quality of 40 and higher ? Is this method better than selecting out with sed everytime?

Answer 1

[Chris Miller]

IIRC, you'll want to set the minimum mapping quality using the q flag:

breakdancer_max <analysis.config>

Options: 
       -o STRING       operate on a single chromosome [all chromosome]
       -s INT          minimum length of a region [7]
       -c INT          cutoff in unit of standard deviation [3]
       -m INT          maximum SV size [1000000000]
       -q INT          minimum alternative mapping quality [35]
       -r INT          minimum number of read pairs required to establish a connection [2]
       -x INT          maximum threshold of haploid sequence coverage for regions to be ignored [1000]
       -b INT          buffer size for building connection [100]
       -t              only detect transchromosomal rearrangement, by default off
       -d STRING       prefix of fastq files that SV supporting reads will be saved by library
       -g STRING       dump SVs and supporting reads in BED format for GBrowse
       -l              analyze Illumina long insert (mate-pair) library
       -a              print out copy number by bam file rather than library, by default on
       -h              print out Allele Frequency column, by default off
       -y INT          output score filter [30]

That said, I'm not sure that changing the mapping quality from 35 to 40 is going to result in a big difference in your results.