Hi Sarah,

Sorry for the late reply. The error is related to a bug we're currently looking into. If you run it again, does it have this error consistently or does it only happen once in a while?

Thanks,

K

On Thu, Jul 7, 2011 at 12:11 PM, Sarah Diehl <diehl@immunbio.mpg.de> wrote:
Hi everybody,

when I run the build_profile_indexes.py script in the scripts/tools/annotation_profiler/ directory on hg19 (downloaded yesterday from UCSC), I get the following error:

Created table dir (profiled_annotations/hg19/wgEncodeOpenChromFaireGlioblaBaseOverlapSignal).
['fileName']
Table wgEncodeOpenChromFaireGlioblaBaseOverlapSignal (/galaxy/galaxy_data/ucsc_data/hg19/wgEncodeOpenChromFaireGlioblaBaseOverlapSignal.sql) does not appear to have a chromosome, a start, or a stop.
Removing empty table (wgEncodeOpenChromFaireGlioblaBaseOverlapSignal) directory (profiled_annotations/hg19/wgEncodeOpenChromFaireGlioblaBaseOverlapSignal).
Created table dir (profiled_annotations/hg19/wgEncodeCshlShortRnaSeqK562ChromatinShortTransfrags).
['bin', 'chrom', 'chromStart', 'chromEnd', 'name', 'score', 'strand', 'length', 'numUnique', 'numReads', 'minSeqCount', 'maxSeqCount', 'aveSeqCount', 'firstSeqCount', 'medSeqCount', 'thirdSeqCount', 'minReadCount', 'maxReadCount', 'aveReadCount', 'firstReadCount', 'medReadCount', 'thirdReadCount', 'numRegions', 'regStart', 'regLength', 'seqCount', 'regCount', 'sumCount', 'KEY', 'KEY']
Traceback (most recent call last):
 File "/galaxy/galaxy_backup/galaxy-dist/scripts/tools/annotation_profiler/build_profile_indexes.py", line 338, in <module>
   if __name__ == "__main__": __main__()
 File "/galaxy/galaxy_backup/galaxy-dist/scripts/tools/annotation_profiler/build_profile_indexes.py", line 310, in __main__
   bitset_dict[ chrom ].set_range(  start, end - start  )
 File "bitset.pyx", line 128, in bx.bitset.BitSet.set_range (lib/bx/bitset.c:1130)
 File "bitset.pyx", line 93, in bx.bitset.b_check_range_count (lib/bx/bitset.c:656)
IndexError: End 16573 is larger than the size of this BitSet (16571).


Everything worked fine with mm9 and dm3.

Since the script wasn't changed in the last year, I suspect it has to do with the UCSC data.

We don't really use hg19, so I'm fine with just leaving it out for now, but I wanted to report the error anyway.

Best regards,
Sarah
___________________________________________________________
Please keep all replies on the list by using "reply all"
in your mail client.  To manage your subscriptions to this
and other Galaxy lists, please use the interface at:

 http://lists.bx.psu.edu/