Single-end reads in BAM cause error?

[seandavi]

I know there has been no official release, so feel free to ignore. However, I have some data that have gone through adapter trimming resulting in some single-end reads in an otherwise paired-end genomic bam file. These have a flag of 0 and a mate chromosome of '*'. This results in an error and ends the workflow execution. Using manta-0.20.1 in somatic calling mode, I see:

SVbreakend has unknown chromosome id in candidate sv
...
remote bam record: NONE

[ctsa]

Hi Sean,

Thanks for the report. Is the number of single end reads in this case small enough that it would be okay if manta just ignored these? I can make a fix to ignore this case at a minimum, ideally we could bring these reads back in for assembly/split read scoring but we might have to break that out as a second ticket.

If you can point to a public dataset that exhibits this problem I can add it to our test suite.

-Chris

From: Sean Davis [mailto:notifications@github.com]
Sent: Sunday, June 08, 2014 07:31
To: StructuralVariants/manta
Subject: [manta] Single-end reads in BAM cause error? (#1)

I know there has been no official release, so feel free to ignore. However, I have some data that have gone through adapter trimming resulting in some single-end reads. These have a flag of 0 and a mate chromosome of '*'. This results in an error and ends the workflow execution. Using manta-0.20.1, I see:

SVbreakend has unknown chromosome id in candidate sv

...

remote bam record: NONE

—
Reply to this email directly or view it on GitHubhttps://github.com//issues/1.

[seandavi]

Yep. Ignoring for now is fine. It would be great to include in assembly/split-reads scoring, but I am mainly interested in giving manta a try given your nice results in the DREAM SV challenge. I don't have a public dataset in mind that exhibits the problem, but if you need a BAM file with single-end reads, I can probably prepare one....

Thanks for looking into this and for your continued devotion to open-sourcing your work.

[ctsa]

Sent a trial fix your way... if that doesn't work I cherry-picked the patch back onto this branch

https://github.com/StructuralVariants/manta/tree/MANTA-183-backport

[ctsa]

Fix available on v0.23.0+ now.

[seandavi]

Thanks, Chris.

On Wed, Jun 10, 2015 at 6:20 PM, Chris Saunders notifications@github.com
wrote:

Fix available on v0.23.0+ now.

—
Reply to this email directly or view it on GitHub
#1 (comment).