multireads option

[MingL196]

Hi,

I've been been having trouble getting Rcount-distribute to run Rcount-multireads output:
I took a bam file, and ran it through Rcount-multireads.
I then took the weighted result and tried running it through Rcount-distribute with the "Use multiple alignments" option checked.

It gave me the error:
"/usr/local/include/seqan/bam_io/read_bam.h:201 Assertion failed : remainingBytes > 4 was: 1 <= 4
Aborted (core dumped)"

Do you have any idea what went wrong?

Attached below is a portion of the bam file (it exhibits the same error behavior) (tiny.bam),
the Rcount-multireads weighted bam file (tiny_multi_weighted.bam),
and the project settings file (temp.xml).

Many thanks,
Ming

tiny.bam.gz
temp.xml.gz
tiny_multi_weighted.bam.gz

[MWSchmid]

Hi Ming

How much RAM do you have? Can you check in the activity monitor/task manager if you run out of it? The reason is that multireads creates a vector as long as a the chromosome - so with little RAM it will fail. I will otherwise have a look at the error in a few days - I'm currently a bit busy -.-

Best regards,

Marc

[MingL196]

No problem.

I have 4g RAM.
The error didn't occur when I ran Rcount-multireads.
It occured when I fed the Rcount-multiread output into Rcount-distribute.

Thank you for your time,
Ming

[MingL196]

.

[MWSchmid]

Hi Ming

I get the same error on my MacOS - I don't know yet what's the reason (the different read lengths, the Ns in the sequences, and the missing index are all not making a difference). Hm - the only quick check coming to my mind would be to use Subread instead of Lifescope (Subread can also align color space reads)... Sorry for not having a better solution and taking quite a long time for it.

Best regards,

Marc