[Debian-med-packaging] Bug#779296: trimmomatic: directory of adapters missing from package

[Michael Crusoe]

On Thu, Feb 26, 2015 at 4:26 PM Andreas Tille <andreas at an3as.eu> wrote:

> Hi Michael,
>
> On Thu, Feb 26, 2015 at 09:16:07PM +0000, Michael Crusoe wrote:
> > > Since we are using a wrapper anyway I have set this inside the wrapper
> > > and thus do not see any need for README.Debian.  If you agree, please
> > > have a look at
> > >
> > >    https://people.debian.org/~tille/packages/trimmomatic/
> > >
> > > and please test the package whether I have interpreted your links
> > > correctly and the package can be uploaded as is.  If you could provide
> > > a test using the adapters this would be more than welcome.  Our goal
> > > should be tro get at least 80% of our packages autopkgtest-able.
> > >
> >
> > Trimmomatic uses a nonstandard argument parser. "ILLUMINACLIP" is command
> > line option, not an environmental variable. Adding `export
> > ILLUMINACLIP=/usr/share/${ILLUMINACLIP}/adapters` to the wrapper script
> > will have no effect.
> >
> > Here is an example command line
> >
> > TrimmomaticSE --phred33 input.fq.gz output.fq.gz
> > ILLUMINACLIP:/usr/share/trimmomatic/TruSeq3-SE:2:30:10 LEADING:3
> TRAILING:3
> > SLIDINGWINDOW:4:15 MINLEN:36
>
> OK.  Would you mind having a look at
>
>    https://people.debian.org/~tille/packages/trimmomatic/
>
> again?  Could you confirm that this package will close the bug and
> provides what users expect?
>

The command I gave you was from their webpage; alas it contains two small
errors: should be `-phred33` (single dash) and the adapter file needs to be
a correct path; the file extension is missing.

TrimmomaticSE -phred33 input.fq.gz output.fq.gz
ILLUMINACLIP:/usr/share/trimmomatic/TruSeq3-SE.fa:2:30:10
LEADING:3 TRAILING:3 SLIDINGWINDOW:4:15 MINLEN:36

This runs successfully on my system with local data.


> > However no FASTQ format sequence files are provided in the package so I
> > can't write a test case.
>
> Well, we actually can add our own to debian/tests, can't we?
>

Right. I've tracked down the sequences they used in their paper but I can't
find license nor copyright information:
http://www.ebi.ac.uk/ena/data/view/SRX131047

While these files are rather large (113MiB) I could generate a subset for
testing purposes.


>
> Kind regards
>
>         Andreas.
>
> --
> http://fam-tille.de
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.alioth.debian.org/pipermail/debian-med-packaging/attachments/20150226/bdbfcb33/attachment.html>