Am using R to do some clustering using agnes(from cluster package) my input is a dissimilarity table. i would like to attach some labels to the resulting tree. Any ideas about the way to do that?
if you have the original matrix from which you calculated the dissim object (of class dist), then the row names of the original matrix will be carreid over and will show up when you plot the dendrogram. If you don't have the original matrix, just add an attribute called "Labels" to the dissimilarity object
- Rajarshi Guha
Thanks Rajarsh! but it does not seem to do that. Now i have the dissimilarity, what the syntax for adding the labels attribute? i could be getting it wrong
- george
Have any of you guys successfully used RSPerl to pull R functionality into Perl. Ultimately I guess it is more for convenience as you can always pick up R out files, but that seems unnecessarily painful. I have it working in fits and starts, big issues with the RReferences perl module at the moment. Any help much appreciated !
Well there are several not so well publicised bugs in the RReferences Perl module that comes with the latest release v0.92. Comments in the wrong place, calls in methods to overloaded but not defined functions. Not cool.
- Ian Simpson
Haven't used RSPerl, but did try another R-Perl bridge a few years ago. Forget the name, but it was equally bad code quality. Switch to another language and try RSRuby or RPy instead ;-)
- Neil Saunders
Thanks Neil, it's certainly looking that way, now I have my excuse to branch out (excuse the pun !)
- Ian Simpson
I've never managed to get any of the Perl/R interfaces working reliably. I'm generally not bothered about getting R data into a Perl data-structure though, I usually just use Perl for sticking scripts together, so I tend to leave the actual data as RData files. PL/R (http://www.joeconway.com/plr...) looks quite cool (although I haven't tried it) - gives you the ability to...
more...
- Cass Johnston
Hi Cass, thanks for that. Well I have it working, but not very happy with it. I think part of the problem is that it seems to be mainly designed for people to do R->Perl rather than the other way around. I too have been doing it the way you do, but wanted something more satisfying and efficient (maybe I am just asking too much !). I will have a look at PL/R thanks for the spot.
- Ian Simpson
Indeed, I've already seen the review of it on CPAN..Ouch. It has been picked up unofficially by another coder who has released a new partial version upgrade v0.03 ? , but the phrase "lost cause" might be appropriate here.
- Ian Simpson
I don't know of any tools specifically for sequence assembly in BioC but the people who would know live here: https://stat.ethz.ch/mailman... which is a list for BioC handling of sequence data from Solexa/SOLiD/454 etc.
- Daniel Swan
Dear Swan., Thanks for the information. I keep getting in touch with those guys. Regards, Purna
- Purnachander G
from email
Do let us know how you find it. I love the idea and see the potential but haven't been able to justify using it in any projects yet.
- Rob Syme
Saw this a while ago but looked inactive; seems to have been updated Feb 2009. Hopefully better than the Perl R wrapper, which basically just ran R externally and tried to parse the output, as I remember.
- Neil Saunders
I've used RSRuby in a few projects and been generally satisfied. It doesn't actually do that much, but it's nice to pull all your methods together into a single script sometimes.
- Chris Miller
Is this worth checking out? I got some spammy emails from them, which made me hesitant to try it, even though they were offering a free one year trial for academics.
- Chris Miller
If you're tied to Word and liked the idea of Sweave (i.e., R code in LaTeX) documents, this seem like a useful approach
- Rajarshi Guha
If you're tied to Word and liked the idea of Sweave (i.e., R code in LaTeX) documents, this seem like a useful approach
- Rajarshi Guha