Re: RFS: pigx-rnaseq
Hi Andreas,
Thank you for your feedback.
On 24.09.20 14:54, Andreas Tille wrote:
> Hi Steffen,
>
> thanks a lot for your continuous work on this.
>
> On Thu, Sep 24, 2020 at 01:38:46PM +0200, Steffen Möller wrote:
>> It was a missing JavaScript build dependency - pigx-rnaseq now builds in
>> cowbuilder.
>>
>> https://salsa.debian.org/med-team/pigx-rnaseq
>>
>> Please kindly have a look.
> My way to check missing copyrights is to grep for the term copyright and
> exclude everything that does not need to be mentioned and also what is
> actually mentioned. In this case it boils down to:
>
> (master) $ LC_ALL=C grep -Ri copyright | grep -v -e ^debian -e ^\.git -e "^Binary file" -e '^LICENSE' -e 'Bora Uyar' -e 'Jonathan Ronen' -e 'Ricardo Wurmus'
> m4/ax_python_module.m4:# Copyright (c) 2008 Andrew Collier
> m4/ax_python_module.m4:# permitted in any medium without royalty provided the copyright notice
> m4/ax_compare_version.m4:# Copyright (c) 2008 Tim Toolan <toolan@ele.uri.edu>
> m4/ax_compare_version.m4:# permitted in any medium without royalty provided the copyright notice
> m4/ax_r_package.m4:# permitted in any medium without royalty provided the copyright notice
>
> which means the files in m4/ either need to be mentioned in d/copyright
> or these can be replaced somehow by Debian packaged files and thus
> beeing excluded from the tarball. Since I did not dived deeply into the
> package I have no idea whether the latter simpler solution can be
> applied and thus I leave it to you to do whatever you consider sensible
> to make ftpmaster accept this.
Argh. Thanks. Fixed.
>> I am rather happy about it. Correct me if I
>> am wrong, but from what I oberve it is the first RNAseq workflow that we
>> have as a package in Debian. I suggest we prepare a Debian News
>> "article" that could then go out once the package is accepted.
> Feel free to draft such an article!
Will do.
>> I web-searched a bit and found
>> https://github.com/urmi-21/COVID-19-RNA-Seq-datasets. One of the bulk
>> data sets in there should then serve as a tutorial, I suppose.
> Please mention this in README.Debian.
Er - no. This does not belong there. It is just a package. But we are
getting closer and closer to the data front. We need a few examples and
a web site to inspect the results, and maybe an extra package to collect
the code. But then, how Debian presents itself as a mediator to the
SARS-CoV-2 data I suggest we pose as a challenge out to the world. All
the many upstreams that have contributed a library to our distribution
have certainly some idea about what we should be doing on what kind of
samples.
I suggest we start discussion this in a separate thread and then reach
out also to the Debian Cloud folks. My dream scenario is that we have
one reference setup for DM/DDs. The cloud resources mimic (prefered) or
copy that reference setup for the different cloud infrastructures. The
research community is then free to use that (just by mounting the) with
the packages we provide or they use their alternative setups - from
which we can then learn again.
> BTW, I'm not sure what you mean in debian/TODO. Is this an open todo
> item or not? I also assume by "Auto tests" you mean upstream build time
> tests. Is this correct?
I was working on three different machines. On one of these I have
already removed d/TODO - and not pushed at is seems. I had meant the
build time tests. Which are intense.
One of the patches I forwarded upstream
(https://github.com/BIMSBbioinfo/pigx_rnaseq/pull/79).
Please kindly revisit the package.
Steffen
Reply to: