Rsubread stopping half-way through align
9
0
Entering edit mode
@campbell1933-11092
Last seen 8.4 years ago

Hi

I've been using Rsubread with a range of new and old FASTQ files and it has been working fine. The latest batch seems to be having some problems and for some of the files I get the following error after it crashes

Traceback:

 1: .C("R_align_wrapper", as.integer(n), as.character(cmd), PACKAGE = "Rsubread")

 2: align(index = "reference_index", minFragLength = 35, nthreads = 30,     readfile1 = "RS-02239140_N3_DMSO_GFP_R1.fastq.gz", output_file = "RWPE.GFP.DMSO.3.BAM")

I think i saw an earlier post about 10 months ago with a similar issue but couldn't see a resolution.

I'm using Rsubread_1.22.2

Anyone got any thoughts?

thanks

 

rsubread • 4.0k views
ADD COMMENT
0
Entering edit mode

Your post shows a function call but doesn't show any error message. What error message did you get? What you mean by "it crashes"? 

AFAIK there are no unresolved issues with Rsubread unexpectedly stopping.

ADD REPLY
0
Entering edit mode

That was the error message and then it gives the options 1-4 for how to quit out of R. By crashes then i mean it doesn't do anything else. Maybe that's not the right word?

ADD REPLY
0
Entering edit mode

The output from traceback is not an error message, so you are keeping us completely in the dark as to what may have gone wrong for you.

ADD REPLY
0
Entering edit mode

Could you provide your commands and also the full screen output?

ADD REPLY
0
Entering edit mode

OK will do. I'm traveling so this will take a couple of days. thanks

ADD REPLY
0
Entering edit mode

The samples divided and run on Hiseq2500 v3 chemistry, high output flowcell, 100SE. The FASTQ files were then catenated.

R version 3.3.0 (2016-05-03) -- "Supposedly Educational" being run on a computer cluster.

In Rsubread

>align(index="reference_index", minFragLength=35, nthreads=30, readfile1="RS-02239142_N4_DMSO_input_R1.fastq.gz", output_file="RWPE.Inp.DMSO.1.BAM")

||                       Threads : 30                                         ||

||                  Phred offset : 33                                         ||

||                     Min votes : 3 / 10                                     ||

||    Maximum allowed mismatches : 3                                          ||

||   Maximum allowed indel bases : 5                                          ||

|| # of best alignments reported : 1                                          ||

||                Unique mapping : yes                                        ||

||                                                                            ||

\===================== http://subread.sourceforge.net/ ======================//

 

|| The input file contains base space reads.                                  ||

|| The range of Phred scores observed in the data is [2,41] 

 

and then

 

|| The range of Phred scores observed in the data is [2,41]                   ||

....it runs for a while.....

 *** caught segfault ***

address 0x7f3a5ee09fd0, cause 'invalid permissions'

 

Traceback:

 1: .C("R_align_wrapper", as.integer(n), as.character(cmd), PACKAGE = "Rsubread")

 2: align(index = "reference_index", minFragLength = 35, nthreads = 30,     readfile1 = "RS-02239142_N4_DMSO_input_R1.fastq.gz", output_file = "RWPE.Inp.DMSO.1.BAM")

 

 

ADD REPLY
0
Entering edit mode

I suspect the problem might be related to the concatenation of your fastq files. Can you try to run align() on one of the fastq files before you merge them?

ADD REPLY
0
Entering edit mode

Yes, that was our thought too, but the non-concatentated file also failed

 

ADD REPLY
0
Entering edit mode

Could you provide us one of your fastq files in question so we can take a close look? Which reference genome did you use in your mapping?

ADD REPLY
1
Entering edit mode
Wei Shi ★ 3.6k
@wei-shi-2183
Last seen 9 days ago
Australia/Melbourne

Thanks for sending through the data. But we could not reproduce the problem you encountered. Both fastq files were successfully aligned on a linux computer.

Have you checked if you have enough disk space? Could you also provide your session info?

 

ADD COMMENT
0
Entering edit mode
@campbell1933-11092
Last seen 8.4 years ago

 

We mapped to Hg19

What's the best way to share a fastq file with you?

ADD COMMENT
0
Entering edit mode
Wei Shi ★ 3.6k
@wei-shi-2183
Last seen 9 days ago
Australia/Melbourne

Could you transfer the data via CloudStor which is fast and free to academic users?

https://cloudstor.aarnet.edu.au/sender/

ADD COMMENT
0
Entering edit mode
@campbell1933-11092
Last seen 8.4 years ago

send you a link to a google drive folder with two fastq files in it. One worked and one didn't. There's also a txt file of the readout from Rsubread for both files.

 

ADD COMMENT
0
Entering edit mode
@campbell1933-11092
Last seen 8.4 years ago

Hi - did you get the link to share the folder through google drive?

 

ADD COMMENT
0
Entering edit mode
@campbell1933-11092
Last seen 8.4 years ago

ok 

yes, we think it's something to do with disk space too! 

Ok, i'll get session info

ADD COMMENT
0
Entering edit mode
@campbell1933-11092
Last seen 8.4 years ago

So, to date, we've been doing as a fisbatch job and then tried modifying the memory allocation

fisbatch --nodes=1 --ntasks-per-node=16 --constraint=CPU-E5-2660

But still seem to be running into the same issue.  

I'm guessing fisbatch won't work always (because of memory allocation issue) and so it will have to  be a slurm job?

 

ADD COMMENT
0
Entering edit mode

I think you'd better seek help from your system admin.
 

ADD REPLY
0
Entering edit mode

ok - thanks - yes it sounds like a system admin issue

 

ADD REPLY
0
Entering edit mode
@campbell1933-11092
Last seen 8.4 years ago

But, one final question - which version of R were you using?

tx

ADD COMMENT
0
Entering edit mode

R 3.3.0 - the same version as you use.

ADD REPLY
0
Entering edit mode
@campbell1933-11092
Last seen 8.4 years ago

thanks.

 

ADD COMMENT

Login before adding your answer.

Traffic: 383 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6