Hi Holger,
I went to the respective file and found that the line which defines seq_file is commented out: (l.38 in def check_seq_file()) ## seq_file = "%s/alignseq.loc" % GALAXY_DATA_INDEX_DIR
This seems to be a bug in the current version of the file.
There is clearly something amiss. In the galaxy-dist source, this line is not commented: https://bitbucket.org/galaxy/galaxy-dist/src/b258de1e6cea/tools/extract/extr...
Removing the comment, the script tries to check for sequence entries in alignseq.loc, which I left empty before, since I didn't need aligned sequences in galaxy until now. Of course this results in another error:
'No sequences are available for 'hg19', request them by reporting this error.'
This is the correct behavior.
I just wanted to raise the question if this dependency is right, wouldn't one rather like to check for the respective build in faseq.loc (unfortunately the file format is different, it doesn't contain the seq in the first column).
Yes, faseq.loc should be used. The use of alignseq.loc is a historical artifact that we haven't fixed yet. If you're inclined to fix it, we'd be happy to incorporate the changes into the code base. Best, J.