A few weeks ago there was some talk about a 'heatmap' function that would plot a heat map with two dendrograms to show clustering (after hclust). I mentioned then that I had been cooking something up. I feel like it's ready to give to the community to use, test, break, etc. Which leads me to my question: How does one go about contributing a function like this, whose singular purpose doesn't warrant a package of its own? Some packages of Bioconductor seem like they might accommodate it now or in the future, such as geneplotter, genecluster (dev). Should I email the maintainers of those packages and try to convince them to wedge in my little function? Any advice? Thanks, Mark Wilkinson Informatics Analyst St. Jude Children's Research Hospital Department of Pharmaceutical Sciences The opinions expressed here are my own and do not necessarily represent those of St. Jude Children's Research Hospital.

On Thu, 13 Mar 2003, Wilkinson, Mark wrote: > mentioned then that I had been cooking something up. I feel like it's ready > to give to the community to use, test, break, etc. Which leads me to my Excellent! > question: How does one go about contributing a function like this, whose > singular purpose doesn't warrant a package of its own? Some packages of > Bioconductor seem like they might accommodate it now or in the future, such > as geneplotter, genecluster (dev). Should I email the maintainers of those > packages and try to convince them to wedge in my little function? Any > advice? I would say that it is probably up to you on how to proceed. One way is to do exactly what you said, contact the maintainer(s) of the package that you feel is appropriate for the function to live in - and then they can take it from there. The other way I'd suggest going about it is to just post here about your function (perhaps even posting the code), and a discussion can be raised about it. This has the affect of also allowing the general Bioconductor public to play with the functionality at that stage as well. -Jeff

>>>>> "MarkW" == Wilkinson, Mark <mark.wilkinson@stjude.org> >>>>> on Thu, 13 Mar 2003 10:10:36 -0600 writes: MarkW> A few weeks ago there was some talk about a 'heatmap' MarkW> function that would plot a heat map with two MarkW> dendrograms to show clustering (after hclust). I MarkW> mentioned then that I had been cooking something up. MarkW> I feel like it's ready to give to the community to MarkW> use, test, break, etc. Which leads me to my MarkW> question: How does one go about contributing a MarkW> function like this, whose singular purpose doesn't MarkW> warrant a package of its own? Some packages of MarkW> Bioconductor seem like they might accommodate it now MarkW> or in the future, such as geneplotter, genecluster MarkW> (dev). Should I email the maintainers of those MarkW> packages and try to convince them to wedge in my MarkW> little function? Any advice? heatmap() will be a function in the next version of R, thanks to Andy Liaw [post to here] and Robert Gentleman. (and myself). Robert Gentleman (mostly) and I (somewhat) have added functionality to the "dendrogram" class in R's mva package. heatmap() builds on that. In general, e-mailing package maintainers is a good idea, or even posting to the "R-devel" mailing list -- which is there for the purpose of discussing future enhancements for R. Depending on the topic, you might get more interested parties than with the bioconductor list. Martin Maechler <maechler@stat.math.ethz.ch> http://stat.ethz.ch/~maechler/ Seminar fuer Statistik, ETH-Zentrum LEO C16 Leonhardstr. 27 ETH (Federal Inst. Technology) 8092 Zurich SWITZERLAND phone: x-41-1-632-3408 fax: ...-1228 <><

<<clusterplot_0.1.tar.gz>> Oh well. Should anyone be interested in my attempt, I'm attaching the code (in a package, for convenience) anyway. Thanks, Mark Wilkinson Informatics Analyst St. Jude Children's Research Hospital Department of Pharmaceutical Sciences The opinions expressed here are my own and do not necessarily represent those of St. Jude Children's Research Hospital. -----Original Message----- From: Martin Maechler [mailto:maechler@stat.math.ethz.ch] Sent: Thursday, March 13, 2003 11:19 AM To: Wilkinson, Mark Cc: Bioconductor (E-mail) Subject: Re: [BioC] contributing to Bioconductor >>>>> "MarkW" == Wilkinson, Mark <mark.wilkinson@stjude.org> >>>>> on Thu, 13 Mar 2003 10:10:36 -0600 writes: MarkW> A few weeks ago there was some talk about a 'heatmap' MarkW> function that would plot a heat map with two MarkW> dendrograms to show clustering (after hclust). I MarkW> mentioned then that I had been cooking something up. MarkW> I feel like it's ready to give to the community to MarkW> use, test, break, etc. Which leads me to my MarkW> question: How does one go about contributing a MarkW> function like this, whose singular purpose doesn't MarkW> warrant a package of its own? Some packages of MarkW> Bioconductor seem like they might accommodate it now MarkW> or in the future, such as geneplotter, genecluster MarkW> (dev). Should I email the maintainers of those MarkW> packages and try to convince them to wedge in my MarkW> little function? Any advice? heatmap() will be a function in the next version of R, thanks to Andy Liaw [post to here] and Robert Gentleman. (and myself). Robert Gentleman (mostly) and I (somewhat) have added functionality to the "dendrogram" class in R's mva package. heatmap() builds on that. In general, e-mailing package maintainers is a good idea, or even posting to the "R-devel" mailing list -- which is there for the purpose of discussing future enhancements for R. Depending on the topic, you might get more interested parties than with the bioconductor list. Martin Maechler <maechler@stat.math.ethz.ch> http://stat.ethz.ch/~maechler/ Seminar fuer Statistik, ETH-Zentrum LEO C16 Leonhardstr. 27 ETH (Federal Inst. Technology) 8092 Zurich SWITZERLAND phone: x-41-1-632-3408 fax: ...-1228 <>< -------------- next part -------------- A non-text attachment was scrubbed... Name: clusterplot_0.1.tar.gz Type: application/octet-stream Size: 39960 bytes Desc: not available Url : https://www.stat.math.ethz.ch/pipermail/bioconductor/attachments /20030313/90ab754c/clusterplot_0.1.tar-0001.exe