[BioC] DNAString ("GC") -- error in nchar?

[Martin Morgan]

Hi Paul -- If something like (using '%' to mean the shell prompt)

% R --vanilla
> nchar('AT')

works, but

% R
> nchar('AT')

produces an error, then likely you've got something stored in
.Rprofile, .Rdata, or other (see ?Startup) startup file that is
confusing R, e.g., pointing to an old installation.

On the other hand if

% R --vanilla
> nchar('AT')

works but

% R --vanilla
> library(Biostrings)
> nchar('AT')

doesn't, then it points toward Biostrings and it would help to see
traceback() and the complete session. If there's no error ever with R
--vanilla, then it is again likely to be something in one of your
profile / environment / site files.

Martin

Paul Shannon <pshannon at systemsbiology.org> writes:

> Good idea, Martin:  in a clean R, nchar works fine.
>
> Stranger and stranger...
>
>   - Paul
>
>
> On Sep 28, 2007, at 3:19 PM, Martin Morgan wrote:
>
>> I wonder what traceback() says immediately after the error, and
>> whether you get the message in a clean R (R --vanilla from the command
>> line; do not load Biostrings)?
>

-- 
Martin Morgan
Bioconductor / Computational Biology
http://bioconductor.org