[R] Creating new data frame using loop

[Steve Lianoglou]

Hi Andy,

On Jul 27, 2009, at 12:18 PM, Andrew Aldersley wrote:

>
> Hi all,
>
> I sent a request round last week asking for help with using a "for"  
> loop to read and separate a large dataset. The response I got worked  
> great, but now I have another problem with using my loop.
>
> Basically I have a number of different files containing columned  
> data. There are 132 datasets, named such that I have something in  
> the form...
>
> precip_colxxx.txt
>
> ...where xxx is a number ranging from 1 to 132. What I want to do is  
> read in every 13th table and extract the third column, and then  
> place this in a new dataset. The new dataset will thus compose of 11  
> columns of data. I have written the following bit of script to read  
> in every 13th table separately, however I'm not sure how to do the  
> next step of creating a new data frame and "dumping" the third  
> column of my tables into this data frame. Is there are chance I will  
> have to do a nested loop?
>
> for (i in seq(1,120,13)) {
>
> nm <- sprintf('precip_col%03d.txt', i)
>
> precip <- read.table(nm, header=T)
>
> }

You can do this by "building up" your columns into a list, then using  
a combo of do.call and cbind.

For example:

mydata <- list()
for (i in seq(1,120,13)) {
   nm <- sprintf('precip_col%03d.txt', i)
   precip <- read.table(nm, header=T)
   mydata[[i]] <- precip[,3]
}

mydata <- do.call(cbind, mydata)

The first param in do.call is the function you want to call, the  
second param is a *list* of parameters you'd like to pass into the  
function.

-steve

--
Steve Lianoglou
Graduate Student: Computational Systems Biology
   |  Memorial Sloan-Kettering Cancer Center
   |  Weill Medical College of Cornell University
Contact Info: http://cbio.mskcc.org/~lianos/contact