Hi Ulrich --
Certainly msa provides very useful functionality not present in other
Bioconductor packages.
It's not possible for the Bioconductor core team to take on additional
maintenance responsibilities; the best bet is to find someone else in the
community willing and able to maintain the package.
If that person does not step forward, then we can proceed with the deprecation
process.
Martin
On 7/28/20, 6:02 AM, "Bioc-devel on behalf of Ulrich Bodenhofer"
wrote:
Dear Lori, dear Martin, dear Hervé,
[I am mentioning you personally, since you are the Core Team members I
had personal contact with in the past]
dear Bioconductor Core Team,
This is an open letter to you and the Bioconductor community ... my
letter concerns the 'msa' package of which I am the maintainer. You are
probably aware that this package wraps three major multiple sequence
alignment algorithms in one Bioconductor/R package. This standard
functionality had not been available in that way in a Bioconductor
package prior to its release. It is, therefore, frequently used (rank
135 of 1879 packages; currently ~2,000 downloads per month) and the
corresponding application note in 'Bioinformatics'
(http://dx.doi.org/10.1093/bioinformatics/btv494) has gained quite a few
citations on Google Scholar so far (tendency still increasing).
I do not want to bother you with details too much, but just a few words
about what this package is internally: it is a wrapper around three
commonly used algorithms. All of these have been published as research
prototypes, as (crude!) source code for user self-compilation under
Linux/Unix. It was quite an effort to make these three complex libraries
run on all three major platforms (and to get rid of their memory leaks).
While the package has worked smoothly for several years now, due to the
progression of R (and also gcc), some issues with Makefiles and
configure scripts have come up that overwhelm me and my competencies. I
should also mention that I am no longer active in the bioinformatics
field and my students who did most of the implementation work are long
gone. The fact remains that 'msa' currently produces check and build
errors on the Windows platform (both release and devel; further issues
have been brought up by Brian Ripley concerning build errors with gcc 10
on Fedora), and I feel unable to fix them.
Now my question: would the Bioconductor Core Team be willing to take
over the 'msa' package? I am sure there are members who are much more
proficient in dealing with compiler flags, configure scripts and all
that stuff than me. I am also convinced (1) that it makes good sense if
such a crucial functionality like sequence alignment is in the hands of
the Core Team, (2) that it would be a great loss to the Bioconductor
project if the package died, and (3) that the Core Team will have great
ideas for further improvements of the package.
What I can offer:
* That you will have full control over the package and its future. I
will not intervene in any way.
* My long-term support if you have questions regarding the details of
the current version.
What I am asking from you:
* Take good care of the package.
* Otherwise: nothing special, but I would be grateful if the paper
(see link above) would remain to be cited in the package (e.g. in
the vignette and in the CITATION file).
I am eager to hear/read your response, thank you very much in advance!
Best regards,
Ulrich
___
Bioc-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/bioc-devel
___
Bioc-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/bioc-devel
