[BioC] how to add gene name in limma

Gordon Smyth smyth at wehi.EDU.AU
Fri Sep 29 11:52:06 CEST 2006


read.maimages() currently ignores the "annotation" and 
"other.columns" arguments if source="imagene". I'm not sure why it 
should be necessary to set the annotation columns for ImaGene files 
because (as far as I know) the column headings are always the same 
(as you give below) and hence already known to limma, but correctly 
using the "other.columns" argument is definitely on the list of 
features to that I intend to add but haven't done so yet.

Best wishes
Gordon


>Date: Thu, 28 Sep 2006 10:12:58 +0200
>From: "Gaj Stan (BIGCAT)" <Stan.Gaj at BIGCAT.unimaas.nl>
>Subject: Re: [BioC] how to add gene name in limma
>To: "Bing Zhang" <biz at exiqon.com>, "john seers (IFR)"
>         <john.seers at bbsrc.ac.uk>
>Cc: bioconductor at stat.math.ethz.ch
>
>Hi,
>
>Sorry for the delayed response (sickness).
>
>After examining the read.imagene function, I noticed that it extracts
>the following annotation (if you 'forget' to call the
>annotation-parameter) from an Imagene filename:
>
>             RG$genes <- obj[, c("Field", "Meta Row", "Meta Column",
>                 "Row", "Column", "Gene ID")]
>
>It seems that the read.maimages-function does something extra with the
>data retrieved from read.imagenes, so I'm not so sure to go only for the
>read.imagene function (although read.imagenes appears to be creating all
>the necessary information as well (-; ).
>
>I noticed in your previous mail that you wanted to extract an annotation
>column called "Annotation 1" as well. Perhaps if the above works fine,
>you can change adjust the code to grab it as well. A dirty approach, I
>agree, but it could solve your problem!
>
>If the above won't work you can follow a last-resort option where you
>first try to read your datafiles with read.maimages, followed by
>extracting your annotation from a tab-delimited file (containing only
>the annotation information in the correct order) (read.table) and put
>that into the correct class in your RG$genes (as.dataframe).
>
>Hope that these suggestions solves your problem!
>
>Best wishes,
>
>    Stan
>
>
>-----Original Message-----
>From: john seers (IFR) [mailto:john.seers at bbsrc.ac.uk]
>Sent: 26 September 2006 16:11
>To: Gaj Stan (BIGCAT); Bing Zhang
>Cc: bioconductor at stat.math.ethz.ch
>Subject: RE: [BioC] how to add gene name in limma
>
>
>
>Hi
>
>Looking at the read.maimages script/code I think perhaps it does not
>handle the annotation values. It tests for "imagene" and then calls
>"read.imagene" without passing on the annotation values. (I may be wrong
>on this - I am not sure what the "..." does). Anyway - my suggestion is
>to try calling "read.imagene" directly specifying the annotation
>parameter.
>
>
>Code from read.maimages:
>
>    if (source2 == "imagene")
>         return(read.imagene(files = files, path = path, ext = ext,
>             names = names, columns = columns, wt.fun = wt.fun,
>             verbose = verbose, sep = sep, quote = quote, ...))
>
>
>Regards
>
>
>J
>
>
>
>
>
>
>---
>
>John Seers
>Institute of Food Research
>Norwich Research Park
>Colney
>Norwich
>NR4 7UA
>
>
>tel +44 (0)1603 251497
>fax +44 (0)1603 507723
>e-mail john.seers at bbsrc.ac.uk
>e-disclaimer at http://www.ifr.ac.uk/edisclaimer/
>
>Web sites:
>
>www.ifr.ac.uk
>www.foodandhealthnetwork.com
>
>
>-----Original Message-----
>From: Gaj Stan (BIGCAT) [mailto:Stan.Gaj at BIGCAT.unimaas.nl]
>Sent: 25 September 2006 12:17
>To: Bing Zhang
>Cc: bioconductor at stat.math.ethz.ch
>Subject: Re: [BioC] how to add gene name in limma
>
>
>Dear Bing,
>
>You're right to use the annotation-parameter of the read.maimages
>function. E.g.:
>
>         annotation = c("GeneName","Description","SystematicName")
>
>Normally it should work fine on importing any other columns (case
>sensitive!) from your tab-delimited text files into the genelist object.
>If this still does not work, could you supply us the actual code that
>you used?
>
>Best wishes,
>
>    Stan
>
>-----Original Message-----
>From: bioconductor-bounces at stat.math.ethz.ch
>[mailto:bioconductor-bounces at stat.math.ethz.ch] On Behalf Of Bing Zhang
>Sent: 25 September 2006 10:09
>To: bioconductor at stat.math.ethz.ch
>Subject: [BioC] how to add gene name in limma
>
>
>Dear all,
>
>I am using limma to generate a genelist object for imagene file.
>
>when I am using the command "read.maimages()", I want to add the gene
>name into
>
>genelist object. I tried to set annotation and other.columns, but it
>seems
>
>none of them worked.
>
>Could anybody tell me how to do that ?
>
>
>
>thanks in advance
>
>
>bing



More information about the Bioconductor mailing list