Hi there I'm trying to wrap Bioperl's bp_genbank2gff script. This script, by default, produces an output file named the same as the basename of the input file, with .gff added. So e.g. bp_genbank2gff data.gbk produces a file named data.gbk.gff. I'm trying to use the from_work_dir attribute of the <data> tag to pick up this file. However, if I create a tool XML like: <tool id="Genbank2GFF3" name="Genbank to GFF3" version="1.0"> <description>Convert Genbank format to GFF3</description> <command interpreter="python"> bp_genbank2gff $genbank_filename </command> <inputs> <param name="genbank_filename" type="data" format="txt"> <label>Genbank sequence record(s)</label> </param> </inputs> <outputs> <data format="gff3" name="gff3_filename" from_work_dir="${genbank_filename}.gff"> </data> </outputs> <help> </help> </tool> I get an error from Galaxy indicating that it is trying to find a file name $genbank_filename.gff, I.e. the ${genbank_filename} is not expanded. Is there any way to deal with this situation? Of course the other option is to create a tiny wrapper around bp_genbank2gff that takes an output filename argument and does the renaming... Peter