Using DEXSeq count on a sliced bam or sam file does not seem to run properly. The output file reads eg., as follows: ambiguous 0 _ambiguous_readpair_position 0 _empty 9893 _lowaqual 0 _notaligned 0 The sliced file had the hg19 genome database .
There are a few different human reference genome versions are being used with tools and assigned to datasets. These are not always a match for the actual database used (example: Tophat run against
hg19 but the BAM is assigned to the database 'hg19Patch2" - which is not indexed for tools). Other times there is a mismatch between inputs or tool form settings. This is probably the root cause of your recent problems.
The same exact reference genome should be used for an entire analysis. If the mapping was run against
hg19, then that should be set as the database, and all other inputs and tool form options should be based on
Hope that helps! Jen, Galaxy team