Hi All,

A quick question on the indices. Here is the information provided by the sample file:

#Your bowtie_indices.loc file should include an entry per line for each
#index set you have stored. The "file" in the path does not actually
#exist, but it is the prefix for the actual index files. For example:

#hg18canon          hg18   hg18 Canonical   /depot/data2/galaxy/bowtie/hg18/hg18canon
#hg18full           hg18   hg18 Full        /depot/data2/galaxy/bowtie/hg18/hg18full
#/orig/path/hg19    hg19   hg19             /depot/data2/galaxy/bowtie/hg19/hg19

#Note that for backwards compatibility with workflows, the unique ID of
#an entry must be the path that was in the original loc file, because that
#is the value stored in the workflow for that parameter. That is why the
#hg19 entry above looks odd. New genomes can be better-looking.

What exactly is "path that was in the original loc file"? Where do I find this "original loc file"? Are we using a newer version of loc file now?

I got the loc file from the main galaxy server, and I could see things like following:

ce5     ce5     C. elegans (WS180): ce5 /galaxy/data/ce5/bowtie_index/ce5
/galaxy/data/ce6/bowtie_index/ce6       ce6     C. elegans (WS190): ce6 /galaxy/data/ce6/bowtie_index/ce6

Why would ce5 and ce6 appear differently? for ce6, seems the "original loc file path" is the same as the "filebasepath"?

galGal3canon    galGal3 Chicken (Gallus gallus): galGal3 Canonical      /galaxy/data/galGal3/galGal3canon/bowtie_index/galGal3canon
/galaxy/data/galGal3/bowtie_index/galGal3       galGal3 Chicken (Gallus gallus): galGal3 Full   /galaxy/data/galGal3/galGal3full/bowtie_index/galGal3full

but as we could see, for this galGal3, the "original loc file path" is different from the "filebasepath". 

same for mm9:

mm9canon        mm9     Mouse (Mus musculus): mm9 Canonical     /galaxy/data/mm9/mm9canon/bowtie_index/mm9canon
mm9female       mm9     Mouse (Mus musculus): mm9 Canonical Female      /galaxy/data/mm9/mm9female/bowtie_index/mm9female
/galaxy/data/mm9/bowtie_index/mm9       mm9     Mouse (Mus musculus): mm9 Full  /galaxy/data/mm9/mm9full/bowtie_index/mm9full

How did this get determined? My conjecture is that in the older version of galaxy we don't have mm9canon and mm9female, for example, but we had mm9, and the entry was like:

mm9         mm9        mm9        /galaxy/data/mm9/bowtie_index/mm9

but now we are using a newer version of galaxy and mm9 genome was placed in 

 /galaxy/data/mm9/mm9full/bowtie_index/mm9full

so we need to modify this loc file to add the mapping so that we don't have to modify the older workflow that uses mm9? is it just because data is organized differently? that's why for all the newer genomes we don't have that?

I'd appreciate that if someone could please elaborate a bit.

Best Regards,
Rui