TopHat2 run with the error message "The Cluster DRM System has terminated this Job"

Question: TopHat2 run with the error message "The Cluster DRM System has terminated this Job"

3.7 years ago by

lax • 0

United States

lax • 0 wrote:

Hello, I use Galaxy regularly, and used TopHat2 earlier without any problem. However, just a day ago, when I was trying to align 'fastq groomed' files with TopHat2, I was getting an error (after a long run): "The Cluster DRM System has terminated this Job". I learnt it's something to do with the limits (time and memory for a run) set by the Galaxy admin? All my 15 'fastq groomed' files eventually returned with the same error message: "The Cluster DRM System has terminated this Job". As I said, I never had problems earlier with aligning my 'fastq groomed' files with TopHat2.

Just to crosscheck, I was able to align all these 15 'fastq groomed' files with Bowtie2, which is actually the core aligner for the TopHat2. So, I was wondering if this has to do with the limits set by the admin. Whom should I talk to?

Our Galaxy runs on Helix clusters. Please advise me. Thanks in advance.

tophat2 bowtie2 • 1.6k views

ADD COMMENT • link •

modified 3.7 years ago • written 3.7 years ago by lax • 0

3.7 years ago by

lax • 0

United States

lax • 0 wrote:

Hello again,

I reran again one of the the "fastq groomed' files, and this time I got this error message. The error message I got on the same file in my previous run did not contain this message. All I got earlier was: " "The Cluster DRM System has terminated this Job". I am pasting error message from the rerun of the same file down here:

Tool execution generated the following error message:

INFO @ Tue, 14 Jan 2014 18:18:05:

# ARGUMENTS LIST:
# name = MACS_in_Galaxy
# format = SAM
# ChIP-seq file = /spin1/users/galaxy/galaxy/database/files/009/dataset_9342.dat
# control file = /spin1/users/galaxy/galaxy/database/files/009/dataset_9343.dat
# effective genome size = 2.70e+09
# tag size = 25
# band width = 300
# model fold = 32
# pvalue cutoff = 1.00e-05
# Ranges for calculating regional lambda are : peak_region,1000,5000,10000
INFO @ Tue, 14 Jan 2014 18:18:05: #1 read tag files...
INFO @ Tue, 14 Jan 2014 18:18:05: #1 read treatment tags...
Traceback (most recent call last):
File "/data/galaxy/appList/MACS-1.3.7.1/bin/macs", line 273, in <module>
    main()
File "/data/galaxy/appList/MACS-1.3.7.1/bin/macs", line 57, in main
    (treat, control) = load_tag_files_options (options)
File "/data/galaxy/appList/MACS-1.3.7.1/bin/macs", line 252, in load_tag_files_options
    treat = options.build(open2(options.tfile, gzip_flag=options.gzip_flag))
File "/usr/local/python-2.6_x86_64/lib/python2.6/site-packages/MACS/IO/init.py", line 1480, in build_fwtrack
    (chromosome,fpos,strand) = self.__fw_parse_line(thisline)
File "/usr/local/python-2.6_x86_64/lib/python2.6/site-packages/MACS/IO/init.py", line 1500, in __fw_parse_line
    bwflag = int(thisfields[1])
ValueError: invalid literal for int() with base 10: '1:N:0:'

Again, as I mentioned earlier, I was able to align the same 'fastq groomed' file with bowtie2, (which is the core aligner of TopHat2) successfully, without any problem.

Thanks!

ADD COMMENT • link modified 3.7 years ago • written 3.7 years ago by lax • 0

This is a traceback from the MACS peakfinding tool. To fix this pasted error, you can convert your SAM file into a BAM file and then MACS will work correctly with this data (it doesn't like the spaces in the read names).

ADD REPLY • link written 3.7 years ago by Daniel Blankenberg ♦♦ 1.7k

Hello Daniel -Thanks for the suggestion. That's a fastq file (fastq groomed) that I am trying to map with TopHat2. When you say convert my SAM file, do you mean convert my fastq file to unmapped BAM file before using TopHat2 on that unmapped BAM file?

Thanks again!

ADD REPLY • link written 3.7 years ago by lax • 0

I think you may have copy and pasted the wrong traceback,. The text you have under "Tool execution generated the following error message:" is an error thrown from the MACS tool, not TopHat2.

ADD REPLY • link written 3.7 years ago by Daniel Blankenberg ♦♦ 1.7k

Hi Daniel -Thanks again. I just checked back my run, and actually did not copy and pasted the wrong traceback. However, I did hear back from my Galaxy at Helix admin, and it has to do with the memory associated with running TopHat2. Here is the message from the admin: "Galaxy is configured to run tophat2 jobs on c24 nodes. Unfortunately, your tophat2 jobs required more than 24GB of memory, so I reconfigured
Galaxy to submit to 72GB nodes for tophat2 jobs. Please rerun these jobs."

So, I am going to rerun again, and hopefully it should work with 72GB nodes. I'm still not sure why I got the MACS traceback. Probably due to the insufficient memory, the tool just throws random errors?

Thanks again.

Lax

ADD REPLY • link written 3.7 years ago by lax • 0

I can't imagine any way that a TopHat2 job would throw a traceback from the MACS code, certainly not randomly.

If this error actually came from a submitted TopHat2 job run, it is probably worth having a system admin look into some sort of cluster configuration issue that could cause a MACS error from a different job run to appear in the Error info for a TopHat2 run.

Tool execution generated the following error message:

INFO @ Tue, 14 Jan 2014 18:18:05:

# ARGUMENTS LIST:
# name = MACS_in_Galaxy
# format = SAM
# ChIP-seq file = /spin1/users/galaxy/galaxy/database/files/009/dataset_9342.dat
# control file = /spin1/users/galaxy/galaxy/database/files/009/dataset_9343.dat
# effective genome size = 2.70e+09
# tag size = 25
# band width = 300
# model fold = 32
# pvalue cutoff = 1.00e-05
# Ranges for calculating regional lambda are : peak_region,1000,5000,10000
INFO @ Tue, 14 Jan 2014 18:18:05: #1 read tag files...
INFO @ Tue, 14 Jan 2014 18:18:05: #1 read treatment tags...
Traceback (most recent call last):
File "/data/galaxy/appList/MACS-1.3.7.1/bin/macs", line 273, in <module>
    main()
File "/data/galaxy/appList/MACS-1.3.7.1/bin/macs", line 57, in main
    (treat, control) = load_tag_files_options (options)
File "/data/galaxy/appList/MACS-1.3.7.1/bin/macs", line 252, in load_tag_files_options
    treat = options.build(open2(options.tfile, gzip_flag=options.gzip_flag))
File "/usr/local/python-2.6_x86_64/lib/python2.6/site-packages/MACS/IO/__init__.py", line 1480, in build_fwtrack
    (chromosome,fpos,strand) = self.__fw_parse_line(thisline)
File "/usr/local/python-2.6_x86_64/lib/python2.6/site-packages/MACS/IO/__init__.py", line 1500, in __fw_parse_line
    bwflag = int(thisfields[1])
ValueError: invalid literal for int() with base 10: '1:N:0:'

ADD REPLY • link written 3.7 years ago by Daniel Blankenberg ♦♦ 1.7k

Tool execution generated the following error message:

INFO @ Tue, 14 Jan 2014 18:18:05:

Similar posts • Search »