Aedin- I played around with SMD import last summer and created a function that much more intelligently imports data from the SMD. The interesting thing about the "Excel" files you get from the SMD is that they aren't actually Excel files. Instead, they are just text files. R can read them just fine. I unfortunately never got around to adding the function into marrayInput. I haven't tested it since last summer so either the SMD or marrayInput schemas may have changed a bit. I'd be happy to help you debug/fix it if incompatibilities have crept in. You can download it here: http://www.colinsmith.org/~colin/read.SMD2.R It tries to shoehorn as much information from the raw data files as possible into the marray classes. The only thing it doesn't import is the control spots because the SMD doesn't include that information in their raw data files. The input is just a character vector of file names. The files should have all used the same chip layout. -Colin On Feb 24, 2004, at 7:59, Aedin wrote: > Date: Tue, 24 Feb 2004 13:58:45 +0000 > From: Aedin <aedin.culhane@ucd.ie> > Subject: [BioC] Reading data from SMD database > To: bioconductor@stat.math.ethz.ch > > Hi all, > Whats the easiest way to read published data in the SMD database into > R. > > The "raw data" from smd.stanford.edu seems to download as multiple > excel > files, which fail to be read into R if I use read.SMD(). Am I > downloading > the incorrect format file from SMD? > > Thanks so much for any help, > Aedin

Thanks a million Colin, I will try this thanks. Aedin -----Original Message----- From: Colin Smith [mailto:colin@colinsmith.org] Sent: 24 February 2004 16:47 To: Aedin Cc: bioconductor@stat.math.ethz.ch Subject: Re: [BioC] Reading data from SMD database Aedin- I played around with SMD import last summer and created a function that much more intelligently imports data from the SMD. The interesting thing about the "Excel" files you get from the SMD is that they aren't actually Excel files. Instead, they are just text files. R can read them just fine. I unfortunately never got around to adding the function into marrayInput. I haven't tested it since last summer so either the SMD or marrayInput schemas may have changed a bit. I'd be happy to help you debug/fix it if incompatibilities have crept in. You can download it here: http://www.colinsmith.org/~colin/read.SMD2.R It tries to shoehorn as much information from the raw data files as possible into the marray classes. The only thing it doesn't import is the control spots because the SMD doesn't include that information in their raw data files. The input is just a character vector of file names. The files should have all used the same chip layout. -Colin On Feb 24, 2004, at 7:59, Aedin wrote: > Date: Tue, 24 Feb 2004 13:58:45 +0000 > From: Aedin <aedin.culhane@ucd.ie> > Subject: [BioC] Reading data from SMD database > To: bioconductor@stat.math.ethz.ch > > Hi all, > Whats the easiest way to read published data in the SMD database into > R. > > The "raw data" from smd.stanford.edu seems to download as multiple > excel > files, which fail to be read into R if I use read.SMD(). Am I > downloading > the incorrect format file from SMD? > > Thanks so much for any help, > Aedin