[Bioc-devel] I will succeed (hopefully) the installation of Bioconductor 2.0

[Daniel Macedo Lorenzini]

Dear Developing Team,

First I would like to thank you for your effort to develop such a nice
code as Bioconductor.

Unfortunately I have found many difficulties during the installation of
Bioconductor. I have started on Bioconductor to analyse a Nimblegen chip
dataset at probe level. Initially I worked on a Debian etch system, with
R 2.4 installed. In the current stable Bioconductor distro the Nimblegen
data is not accepted. So I tried to install the Bioconductor 2.0, which
needs R 2.5 (I don t understand why). There is no debian package for R
2.5, so I got the R source code. R source doesn t compile because it doesn
t like my xorg installation (I do have the x11-dev package installed). So
I gave up my linux installation (shame on me) and went into the Window$ R
2.5. I got it correctly installed, all the affy and oligo packages I
needed were download and I prepared my Nimblegen chip design library with
commands such as the following:

> library(makePlatformDesign)
> makePDpackage("2004-11-09_Human_60mer_TEST.ndf", "ngsExpression1.xys", manufacturer="nimblegen")

for my surprise I could not install such package with the R CMD INSTALL.
Searching some more I found that Perl and MinGW would be useful. Still
didn t make it. I find nothing else to change.

I wonder why the Bioconductor developing team cannot work on a widely
distributed R version. After one week of work I found out that it will be
easier to come back home (linux), prepare a Perl script to fake Nimblegen
data as Affymatrix data and then work on the stable bioconductor version.

Regards,

Daniel Lorenzini