I'm reporting two problems on the most recent Galaxy (lavender background, http://g2.bx.psu.edu:8889/) (The header says "logged in as jb10@sanger.ac.uk) 1. Problem running overlap: When I ran overlap on the transcriptional regulation composite hits (Transcription Regulation (TR (20060206) [gencode_partitioned])) and a CpG island track, I got the following error message: An error occurred running this job: ERROR: commandline error, substituting {'input2': './database/files/dataset_265.dat', 'input1': './database/files/dataset_282.dat', 'input2_strandCol': '0', 'min': 1, 'dbkey': u'hg17', 'input2_chromCol': '1', 'out_file1': './database/ files/dataset_283.dat', 'input2_startCol': '2 ', 'input2_endCol': '3'} into galaxyOps3 $dbkey -chromCol=$input1_chromCol -startCol= $input1_startCol -stopCol=$input1_endCol -strandCol=$input1_strandCol -chromCol2=$input2_chromCol -startCol2=$input2_startCol -stopCol2= $input2_endCol $input1 $input2 -all -chrom=all -minSize=$min -bed= $out_file1 -> 'input1_chromCol' Other tracks did overlap just fine. 2. Problem with "Select lines that match an expression" I got empty files back when I tried two different things. No error was reported, but the result reports " Select on data 30 empty, format: interval, database: hg17 Info: Illegal output filename. Ross Hardison T. Ming Chu Professor of Biochemistry and Molecular Biology The Pennsylvania State University 304 Wartik Laboratory University Park, PA 16802 e-mail: rch8@psu.edu phone: 814-863-0113 FAX: 814-863-7024
It seems that when the server is shut down and then restarted, that people's histories end up assigned to different people some times. Loging out and Logging in seems to fix it, but a real solution should be found. As for the overlap, one possibility is that TR file is not technically a bed file (it has extra columns and floating point numbers). What track were you overlapping with? Overlaping the TR with itself works. I think the current incarnation of opperations may have issues when the two files have different formats? I haven't really spent any time looking at the select tool, so I can say anything off of my head. Thanks for the bug reports, Dan On Fri, 14 Apr 2006, Ross Hardison wrote:
I'm reporting two problems on the most recent Galaxy (lavender background, http://g2.bx.psu.edu:8889/) (The header says "logged in as jb10@sanger.ac.uk)
1. Problem running overlap: When I ran overlap on the transcriptional regulation composite hits (Transcription Regulation (TR (20060206) [gencode_partitioned])) and a CpG island track, I got the following error message:
An error occurred running this job: ERROR: commandline error, substituting {'input2': './database/files/dataset_265.dat', 'input1': './database/files/dataset_282.dat', 'input2_strandCol': '0', 'min': 1, 'dbkey': u'hg17', 'input2_chromCol': '1', 'out_file1': './database/ files/dataset_283.dat', 'input2_startCol': '2 ', 'input2_endCol': '3'} into galaxyOps3 $dbkey -chromCol=$input1_chromCol -startCol=$input1_startCol -stopCol=$input1_endCol -strandCol=$input1_strandCol -chromCol2=$input2_chromCol -startCol2=$input2_startCol -stopCol2= $input2_endCol $input1 $input2 -all -chrom=all -minSize=$min -bed=$out_file1 -> 'input1_chromCol'
Other tracks did overlap just fine.
2. Problem with "Select lines that match an expression" I got empty files back when I tried two different things. No error was reported, but the result reports " Select on data 30
empty, format: interval, database: hg17 Info: Illegal output filename.
Ross Hardison T. Ming Chu Professor of Biochemistry and Molecular Biology The Pennsylvania State University 304 Wartik Laboratory University Park, PA 16802 e-mail: rch8@psu.edu phone: 814-863-0113 FAX: 814-863-7024
The fact that neither you nor Anton replicated the problem suggested to me that something was wrong with the file as I downloaded it. Indeed, when I downloaded TR again, it works for overlaps. Anton had suggested that I change the name or edit the TR file, but neither of these solved the problem. I could not find a difference in the two downloads of TR - same number of regions, and the "compare" tool did not find anything. Thanks, Ross On Apr 14, 2006, at 2:16 PM, Dan Blankenberg wrote:
It seems that when the server is shut down and then restarted, that people's histories end up assigned to different people some times. Loging out and Logging in seems to fix it, but a real solution should be found.
As for the overlap, one possibility is that TR file is not technically a bed file (it has extra columns and floating point numbers). What track were you overlapping with? Overlaping the TR with itself works. I think the current incarnation of opperations may have issues when the two files have different formats?
I haven't really spent any time looking at the select tool, so I can say anything off of my head.
Thanks for the bug reports,
Dan
On Fri, 14 Apr 2006, Ross Hardison wrote:
I'm reporting two problems on the most recent Galaxy (lavender background, http://g2.bx.psu.edu:8889/) (The header says "logged in as jb10@sanger.ac.uk)
1. Problem running overlap: When I ran overlap on the transcriptional regulation composite hits (Transcription Regulation (TR (20060206) [gencode_partitioned])) and a CpG island track, I got the following error message:
An error occurred running this job: ERROR: commandline error, substituting {'input2': './database/files/dataset_265.dat', 'input1': './database/files/dataset_282.dat', 'input2_strandCol': '0', 'min': 1, 'dbkey': u'hg17', 'input2_chromCol': '1', 'out_file1': './database/ files/dataset_283.dat', 'input2_startCol': '2 ', 'input2_endCol': '3'} into galaxyOps3 $dbkey -chromCol=$input1_chromCol -startCol=$input1_startCol - stopCol=$input1_endCol -strandCol=$input1_strandCol -chromCol2= $input2_chromCol -startCol2=$input2_startCol -stopCol2= $input2_endCol $input1 $input2 -all -chrom=all -minSize=$min -bed= $out_file1 -> 'input1_chromCol'
Other tracks did overlap just fine.
2. Problem with "Select lines that match an expression" I got empty files back when I tried two different things. No error was reported, but the result reports " Select on data 30
empty, format: interval, database: hg17 Info: Illegal output filename.
Ross Hardison T. Ming Chu Professor of Biochemistry and Molecular Biology The Pennsylvania State University 304 Wartik Laboratory University Park, PA 16802 e-mail: rch8@psu.edu phone: 814-863-0113 FAX: 814-863-7024
Ross Hardison T. Ming Chu Professor of Biochemistry and Molecular Biology The Pennsylvania State University 304 Wartik Laboratory University Park, PA 16802 e-mail: rch8@psu.edu phone: 814-863-0113 FAX: 814-863-7024
participants (2)
-
Dan Blankenberg -
Ross Hardison