EXITING becaues of BUG: different annotation status for the same junction while collapsing junctions

[xiong...@gmail.com]

I ran STAR to map RNA-Seq reads (about 30nt length) to C.elegans genome. All other files have successfully mapped, except one of them failed with below error. I've tried both 2.3.1 and 2.4.0a, the error was the same. Anyone have idea how to solve it?

/software/STAR-ENCODE_2014-03-16_2.3.1z1/STAR --runThreadN 10 --readFilesCommand zcat --genomeDir /data/Indices/STAR/C_elegans/WB230/ --readFilesIn "sampleA.fastq.gz" --outFileNamePrefix "sampleA"
Oct 11 20:59:56 ..... Started STAR run
Oct 11 20:59:59 ..... Started mapping

EXITING becaues of BUG: different annotation status for the same junction while collapsing junctions:100817015 991 0 1

Oct 11 21:04:07 ...... FATAL ERROR, exiting

[Alexander Dobin]

Hi,

this is a serious error. To track it, I will need your genome and fastq file. You can send me the whole file, or a small portion of it where you can still see the error.

Cheers

Alex

[Peter Sudmant]

Was there any follow up to this? I'm having the same issue - I'm happy to send you fastqs etc. 

The interesting thing about this bug is it seems to be index specific - in two pass mapping the error only occurred on mapping to the second, extended index. 

EXITING becaues of BUG: different annotation status for the same junction while collapsing junctions

Let me know if I should start a new thread

[Alexander Dobin]

Hi Peter,

this was not followed up, and I have not heard about this problem again.

Have you tried the very latest patch of STAR just in case?

I would need the fastq files, the genome you are mapping to, and your parameters to reproduce and fix the problem.

Cheers

Alex

[Peter Sudmant]

Thanks so much for your follow up Alex!

Actually, I was using version 2.4.0i, I just switched to 2.4.0j and reran everything without any problems! I'm not sure if that's expected or not... but, it's good for me!

If it's at all helpful I can post up my index and fastqs for debugging purpose on your end. Let me know.

Peter

[Alexander Dobin]

Hi Peter,

it would be nice if you can send me fastq and indices, since I am not sure how the problem got fixed in 2.4.0j - it's better if I checked it carefully.

Cheers

Alex

[Peter Sudmant]

Oh, I missed this - ok. Will do. I'll contact you by email