[R] reading tables into R. .

[Wagle, Mugdha]

Hi,
 
The file I am reading is a text file, whose contents are a matrix that has 15 rows and 58 columns. The first row has column names, and the first column has row names, so the format is correct as far as using read.table is concerned. The other values in the table are all float values (numeric). So when I read in the file using data1 <- read.table("HAL001_HAL0015_Signals.txt"), it gets read in as a table, but when I try to manipulate an individual value as follows: data1[2 ,2] <- log(data1[2 ,2]+20) , I get the "object is not subsettable" message. This error occurs when I use R only...Perl is not being used at this point. My script needs to take input from an HTML page (using a CGI interface) which will be the name of the file to be passed onto R as a table. Since read.table isn't working within R itself, I haven't used it for the function call from Perl to R yet. Instead, I have been making repeated function calls using a loop in Perl, to access other R and Bioconductor functions that I need such as t.test and ANOVA. This is very time-consuming, however. I am working in a Linux environment, and using
 R 2.1.0
 
Thanks for any help and suggestions!
 
Mugdha Wagle
 
________________________________

From: Sean Davis [mailto:sdavis2 at mail.nih.gov]
Sent: Fri 6/3/2005 8:03 AM
To: Wagle, Mugdha
Cc: r-help at stat.math.ethz.ch
Subject: Re: [R] reading tables into R. .




On Jun 3, 2005, at 8:40 AM, Wagle, Mugdha wrote:

> I have been using R and Perl. When  I read in a text file using the
> read.table option, and I try to mathematically manipulate the
> individual elements in the table, I keep getting an "object is not
> subsettable" error. If I try to use a different method, it works, but
> takes too much time(basically, I then need to read in values
> individually into R instead of as a 2D array, so the number of
> function calls from Perl to R is very large). Could you suggest
> another method whereby I could read an entire matrix or a file using
> an R function call?
>

Mugdha,

I think you will probably have to be more specific.   Could you give an
example of the data format, the commands you used to load it that
didn't work, and those that did and explain how perl comes into this? 
Also, knowing what OS and version of R you are using is quite helpful.

Sean