another broken RUM job

[safisher]

Below is an error file for another RUM job that crapped out. Note that the error file has lots of control characters. That seems somewhat unexpected. Do you think I might have a corrupted data file?

---

Thu Jan 3 09:57:47 2013 16856 FATAL RUM::Death - Unmatched ) in regex; marked by <-- HERE in m/ 43$,-2>16&7,9%;53072"% > 676&? acvvraegv:gvttevt8fgq=2!7>,2 ,vvjr,106.0=,) <-- HERE 6'9 +%,3&67%&1-:7612& avartartvccetggaatatgtgca/ at /home/fisher/local/src/RUM-Pipeline-v2.0.3_04/bin/../lib/RUM/Sort.pm line 75, line 71437762. Thu Jan 3 09:57:47 2013 6473 FATAL RUM::Death - Error running "perl /home/fisher/local/src/RUM-Pipeline-v2.0.3_04/bin/../bin/rum2quantifications.pl --genes-in /lab/repo/resources/rum2/drosophila/drosophila_modencode_gene_info.txt --unique-in /unsafe14/fisher/e.53/Sample_H4D15/rum.trim/chunks/RUM_Unique.sorted.28 --non-unique-in /unsafe14/fisher/e.53/Sample_H4D15/rum.trim/chunks/RUM_NU.sorted.28 -o /unsafe14/fisher/e.53/Sample_H4D15/rum.trim/chunks/quant.28.tmp.7OAWQ51D -countsonly"

The stderr from that command is

Unmatched ) in regex; marked by <-- HERE in m/ 43$,-2>16&7,9%;53072"% > 676&? acvvraegv:gvttevt8fgq=2!7>,2 ,vvjr,106.0=,) <-- HERE 6'9 +%,3&67%&1-:7612& avartartvccetggaatatgtgca/ at /home/fisher/local/src/RUM-Pipeline-v2.0.3_04/bin/../lib/RUM/Sort.pm line 75, line 71437762.

The error log file Sample_H4D15/rum.trim/log/rum_errors_28.log may have more details. Thu Jan 3 10:18:53 2013 16941 FATAL RUM::Death - Unmatched ) in regex; marked by <-- HERE in m/ 43$,-2>16&7,9%;53072"% > 676&? acvvraegv:gvttevt8fgq=2!7>,2 ,vvjr,106.0=,) <-- HERE 6'9 +%,3&67%&1-:7612& avartartvccetggaatatgtgca/ at /home/fisher/local/src/RUM-Pipeline-v2.0.3_04/bin/../lib/RUM/Sort.pm line 75, line 71437762. Thu Jan 3 10:18:53 2013 16908 FATAL RUM::Death - Error running "perl /home/fisher/local/src/RUM-Pipeline-v2.0.3_04/bin/../bin/rum2quantifications.pl --genes-in /lab/repo/resources/rum2/drosophila/drosophila_modencode_gene_info.txt --unique-in /unsafe14/fisher/e.53/Sample_H4D15/rum.trim/chunks/RUM_Unique.sorted.28 --non-unique-in /unsafe14/fisher/e.53/Sample_H4D15/rum.trim/chunks/RUM_NU.sorted.28 -o /unsafe14/fisher/e.53/Sample_H4D15/rum.trim/chunks/quant.28.tmp.gsXqlkRH -countsonly"

The stderr from that command is

Unmatched ) in regex; marked by <-- HERE in m/ 43$,-2>16&7,9%;53072"% > 676&? acvvraegv:gvttevt8fgq=2!7>,2 ,vvjr,106.0=,) <-- HERE 6'9 +%,3&67%&1-:7612& avartartvccetggaatatgtgca/ at /home/fisher/local/src/RUM-Pipeline-v2.0.3_04/bin/../lib/RUM/Sort.pm line 75, line 71437762.

The error log file Sample_H4D15/rum.trim/log/rum_errors_28.log may have more details. Thu Jan 3 10:40:16 2013 17080 FATAL RUM::Death - Unmatched ) in regex; marked by <-- HERE in m/ 43$,-2>16&7,9%;53072"% > 676&? acvvraegv:gvttevt8fgq=2!7>,2 ,vvjr,106.0=,) <-- HERE 6'9 +%,3&67%&1-:7612& avartartvccetggaatatgtgca/ at /home/fisher/local/src/RUM-Pipeline-v2.0.3_04/bin/../lib/RUM/Sort.pm line 75, line 71437762. Thu Jan 3 10:40:16 2013 17049 FATAL RUM::Death - Error running "perl /home/fisher/local/src/RUM-Pipeline-v2.0.3_04/bin/../bin/rum2quantifications.pl --genes-in /lab/repo/resources/rum2/drosophila/drosophila_modencode_gene_info.txt --unique-in /unsafe14/fisher/e.53/Sample_H4D15/rum.trim/chunks/RUM_Unique.sorted.28 --non-unique-in /unsafe14/fisher/e.53/Sample_H4D15/rum.trim/chunks/RUM_NU.sorted.28 -o /unsafe14/fisher/e.53/Sample_H4D15/rum.trim/chunks/quant.28.tmp.jxPIetp1 -countsonly"

The stderr from that command is

Unmatched ) in regex; marked by <-- HERE in m/ 43$,-2>16&7,9%;53072"% > 676&? acvvraegv:gvttevt8fgq=2!7>,2 ,vvjr,106.0=,) <-- HERE 6'9 +%,3&67%&1-:7612& avartartvccetggaatatgtgca/ at /home/fisher/local/src/RUM-Pipeline-v2.0.3_04/bin/../lib/RUM/Sort.pm line 75, line 71437762.

The error log file Sample_H4D15/rum.trim/log/rum_errors_28.log may have more details.

[mdelaurentis]

I'm almost positive this is due to a corrupt intermediate file. I believe I've seen a similar error before, and it was due to a corrupt file. The error about the unmatched ')' in a regular expression is thrown from the code that compares two chromosome names to each other, which is used in sorting alignments by location. The sort function creates a regular expression based on the chromosome names, and uses that for part of the comparison. In this case, one of the chromosome names looks like it's junk.

The script that failed (rum2quantifications) takes three input files: the gene annotation file in the index, the RUM_Unique.CHUNK file, and the RUM_NU.CHUNK file. My guess is that either the RUM_NU or RUM_Unique file is corrupted in some way.

I don't think I have access to the directory of the failed job (I don't have read permissions on /unsafe14/fisher/e.53/Sample_H4D15/rum.trim/chunks). If you want to give me access to it I'll take a look and see which file is messed up. I think you'll probably just need to rerun this one.

On Fri, Jan 4, 2013 at 2:57 PM, safisher notifications@github.com wrote:

/unsafe14/fisher/e.53/Sample_H4D15/rum.trim/chunks

[safisher]

I added you to the lab group so you should now be able to access that directory.

Thanks, Stephen

On Jan 7, 2013, at 10:25 AM, Mike DeLaurentis notifications@github.com wrote:

I'm almost positive this is due to a corrupt intermediate file. I believe I've seen a similar error before, and it was due to a corrupt file. The error about the unmatched ')' in a regular expression is thrown from the code that compares two chromosome names to each other, which is used in sorting alignments by location. The sort function creates a regular expression based on the chromosome names, and uses that for part of the comparison. In this case, one of the chromosome names looks like it's junk.

The script that failed (rum2quantifications) takes three input files: the gene annotation file in the index, the RUM_Unique.CHUNK file, and the RUM_NU.CHUNK file. My guess is that either the RUM_NU or RUM_Unique file is corrupted in some way.

I don't think I have access to the directory of the failed job (I don't have read permissions on /unsafe14/fisher/e.53/Sample_H4D15/rum.trim/chunks). If you want to give me access to it I'll take a look and see which file is messed up. I think you'll probably just need to rerun this one.

On Fri, Jan 4, 2013 at 2:57 PM, safisher notifications@github.com wrote:

/unsafe14/fisher/e.53/Sample_H4D15/rum.trim/chunks

— Reply to this email directly or view it on GitHub.