false positives from BAM file

[cmccabe]

Is there a tool or script that can help identify false positives from a BAM file? Thanks.

[dpryan]

False-positive what? Alignments? Filter by MAPQ.

[cmccabe]

We use Ion Torrent PGM sequencing and use the aligned hg19 BAM in NextGene. I know that TVC is better at removing false positive variants, presumably because it has access to the SFF file and which contain signal values in flow space and a mapping between sequence. Is there a preferred or optimal way to asses MAPQ in the BMA alignment? Thanks.

[dpryan]

As NextGene is a commercial package, you should ask the company that makes it for suggestions. You'll generally find that the community won't provide any support/feedback for such packages.

[cmccabe]

I have contacted them and explained the issue. I am sure that future versions of the software will get better, but in the meantime I want to ensure that we are providing the best information possible. I am looking for some ideas from community members in developing a perl script to help address this. Some of the false positive calls have great PhreD scores (> 20), with good balance, so I am stumped. Thanks.