[R] Reading in and modifying multiple datasets in a loop

[Uwe Ligges]

On 24.10.2011 23:10, Debs Majumdar wrote:
> Thanks Uwe. This works perfectly.
>
> #######
>
>
> owd<- setwd(pth)
> fls<- list.files(pattern="^chr")
> ufls<- unique(sapply(strsplit(fls, "_"), "[", 1))
> for(i in ufls){
>       of<- strsplit(i, "\\.")[[1]]
>       of<- paste(of[1], tail(of, 1), sep=".")
>       impute2databel(genofile = i,
>                      samplefile = paste(i, "info", sep="_"),
>                      outfile = of,
>                      makeprob=TRUE, old=FALSE)
> }
> setwd(owd)
>
> ####
>
>
> I have a question regarding how strsplit works.
>
> When my files are the following:
>
>          chr1.one.phased.impute2.chunk1
>          chr1.one.phased.impute2.chunk1_info
>          chr1.one.phased.impute2.chunk1_info_by_sample
>          chr1.one.phased.impute2.chunk1_summary
>          chr1.one.phased.impute2.chunk1_warnings
> ufls<- unique(sapply(strsplit(fls, "_"), "[", 1))
>
> This works like a charm.
>
> I have another dataset where the files are
>
>
>          study1_chr1.one.phased.impute2.chunk1
>          study1_chr1.one.phased.impute2.chunk1_info
>          study1_chr1.one.phased.impute2.chunk1_info_by_sample
>          study1_chr1.one.phased.impute2.chunk1_summary
>          study1_chr1.one.phased.impute2.chunk1_warnings
>
> ... and so on.
>
> and I wanted to run the same loop but I was unable to change strsplit so that it will work when the files are names ads above:
>
> I tried
>
> ufls<- unique(sapply(strsplit(fls, "_"), "[", 2))


unique(gsub("(_.*)_.*", "\\1", x))

Should do if there is a first underscore.

Uwe Ligges



> but this knocks off "study1" (modified code below).  What modification do I need to make to make this run:
>
> ####
>
> fls<- list.files(pattern="study1_chr")
> ufls<- unique(sapply(strsplit(fls, "_"), "[", 2))
>
> library(GenABEL)
>
> for(i in ufls){
>       of<- strsplit(i, "\\.")[[1]]
>       of<- paste(of[1], tail(of, 1), sep=".")
>       impute2databel(genofile = i,
>                      samplefile = paste(i, "info", sep="_"),
>                      outfile = of,
>                      makeprob=TRUE, old=FALSE)
>
> }
>
> #####
>
> Thanks,
>
>   Debs
>
>
> ----- Original Message -----
> From: Debs Majumdar<debs_stata at yahoo.com>
> To: "r-help at r-project.org"<r-help at r-project.org>
> Cc:
> Sent: Friday, October 21, 2011 2:32 PM
> Subject: Reading in and modifying multiple datasets in a loop
>
>
>
> Hi,
>
>    I have been given a set of around 300 files where there are 5 files corresponding to each chunk.
>
> E.g. Chunk 1 for chr1 contains these 5 files:
>
>          chr1.one.phased.impute2.chunk1
>          chr1.one.phased.impute2.chunk1_info
>          chr1.one.phased.impute2.chunk1_info_by_sample
>          chr1.one.phased.impute2.chunk1_summary
>          chr1.one.phased.impute2.chunk1_warnings
>
> For chr 1 there are 47 chunks, chr2 has 42 chunks...and it ends at chr22 with 23 chunks.
>
> I am using the DatABEL package to  convert them databel format using the following command:
>
>
> impute2databel(genofile="chr1.one.phased.impute2.chunk1", samplefile="chr1.one.phased.impute2.chunk1_info", outfile="chr1.chunk1", makeprob=TRUE, old=FALSE)
>
> which uses two files per chunk.
>
>
> Is there a way I can automate this so that the code goes through each chunk of each chromosome and does the conversion to databel format.
>
>
> Thanks,
>
>   -Debs
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.