I'd like to get a better understanding of the point of the database/build attribute, and pose the question of when is the appropriate time to have it set? In our case at the Jackson Laboratory, the most common build is NCBI37/MM9. However, the feeling of many folks here, is that this should not be set on our fastq files. The only place we really run into trouble is with cufflinks. If you haven't set the db when you get to cufflinks you'll get an error. Our suggestion is that there should be one of two options: 1) Tophat has the ability to set the database of the output files based on the genome that was selected for alignment. 2) There should be a module that can be plugged into a workflow that would set the database of the file prior to passing the file to cufflinks (or any other two that requires the database attribute to be set). We are curious if anyone else is running into this issue, and how it is being solved. We're thinking about hacking the Tophat wrapper, but I wanted to check with others before I did this. Thanks, Dave