[BioC] problem with GEO site

Saurin D. Jani jani at musc.edu
Tue Mar 29 17:18:25 CEST 2005


> I was wishering if it's passible to do a clustering 
> analysis of this file with R ?

you need to parse this file and make expression set in R. for that you need GEO
parser and below is GEO parser that I wrote some time ago. 

##================================================================
##                               GEO SOFT FILES
##================================================================
# GEO soft file parser(1.0) - Saurin Jani

#-- reading soft file

	softFile <- list.files(,"soft"); # from local directory

	 system("cp *.soft file1.soft");
	 system("grep -on \"ID_REF\" file1.soft > b.txt"); 
	 # put your GEO soft file , b.txt file will be created on your computer 
	 
	 system("grep \"dataset_platform\" file1.soft > d.txt");
	 ln <- as.matrix(readLines("b.txt"));
	 lm <- as.matrix(readLines("d.txt"));

	 system("rm b.txt");
	 system("rm d.txt");
	 system("rm file1.soft");

	 lnX <- as.matrix(unlist(strsplit(ln[2],":")))
	 Skpnum <- as.numeric(lnX[1]);

	 lmX <- as.matrix(unlist(strsplit(lm[1],"=")))
	 chiptype <- trimWhiteSpace(lmX[2]);
	 GDSN <-  softFile;

	 emX <- read.table(softFile,skip = Skpnum,comment.char = "");
	 Colm <- ncol(emX);

	 Rnames <- as.matrix(emX["V1"]);
	 temp_emX <- emX;

	 temp2  <- temp_emX[3:Colm];
	 temp2 <- as.matrix(temp2);
	 rownames(temp2) <- Rnames;

#--making expressiong set out of soft file, soft file has normalized data,so I am 
#---assuming here that this data is also normalized

	esetX <- as.matrix(temp2);
	eset <- new("exprSet", exprs = esetX);


you can use eset for clustering.


Saurin
-- 
|------------------------------------------------
| Saurin Jani,MS
| Statistical and Research Analyst
|
| Department of Cell Biology and Anatomy
| Medical  University of South Carolina (MUSC)
| 173 Ashley Ave
| Charleston,SC - 29407 (US)
| 
| Email: jani at musc.edu
| Phone: (843)792-5483
|------------------------------------------------


Quoting Guillaume Deplaine <guillaume.deplaine at college-de-france.fr>:

> Hello,
> 
>     I found a file on GEO web site. this files was processed with MASS 4 
> until normalization. I was wishering if it's passible to do a clustering 
> analysis of this file with R ?
> 
>     My second question is if it's possible to retrieve raw data of this file
> 
> processed with MASS 4?
> 
> Thanks for your answer
> 
> _______________________________________________
> Bioconductor mailing list
> Bioconductor at stat.math.ethz.ch
> https://stat.ethz.ch/mailman/listinfo/bioconductor
> 
>



More information about the Bioconductor mailing list