[BioC] text.to.search problem in read.maimages

[Park-Ng, Zaneta]

Just a wee note that I think that Erik's original column names may have
been correct for his .gpr files.  I have the same column names in some
genepix files that I have, and so his original code runs fine for my
data (albeit that the warning message is obtained).  

My .gpr files come from a GenePix 4200A machine scanned using GenePix
Pro 6.0.  There may have been a change in the column names that GenePix
produces between versions?

Cheers,
Zaneta


-----Original Message-----
From: bioconductor-bounces at stat.math.ethz.ch
[mailto:bioconductor-bounces at stat.math.ethz.ch] On Behalf Of Gordon
Smyth
Sent: Tuesday, 1 May 2007 3:56 p.m.
To: Erik Dopman
Cc: BioC Mailing List
Subject: [BioC] text.to.search problem in read.maimages

Dear Erik,

The error you're receiving is almost certainly due to the fact that 
your function filter() refers to columns which don't exist in GenePix 
data files. Your problem therefore has nothing to do with limma 
versions. You are using the columns "Median of Ratios (635/532)" and 
"Sum of Medians (635/532)", and these column headings don't exist in 
any GenePix file that I've seen. Check the files for the actual 
names. I suspect that you want "Median of Ratios (Ratio/3)" and "Sum 
of Medians".

The warning message you quote actually doesn't have any affect on the 
results. You can ignore it. It does indicate something that I'll fix, 
but it is only a warning, not an error.

Best wishes
Gordon

>[BioC] text.to.search problem in read.maimages
>Erik Dopman ebd5 at cornell.edu
>Wed Apr 25 16:20:21 CEST 2007
>
>Dear List:
>
>I am running R v 2.4.1 and limma v 2.9.17, both of which were
>recently installed from older versions. Since making the update to
>newer versions, I am now unable to run my weight function in the
>read.maimages function. The error that I am receiving is:
>
>Warning message:
>the condition has length > 1 and only the first element will be used
>in: if (text.to.search != "") for (i in 1:ncn) {
>
>Has anybody run into this problem before? My overly redundant filter
>function and scan statement is below.
>
>Thanks,
>Erik
>
>
>filter <- function(x) {
>         ifforeground <- (
>                 (x[,"% > B635+2SD"] > 70 | x[,"% > B532+2SD"] > 70) &
>                 (x[,"F635 Median - B635"] > 4 * x[,"B635 SD"]) &
>                 (x[,"F532 Median - B532"] > 4 * x[,"B532 SD"]) &
>                 x[,"F635 % Sat."] < 30 & x[,"F532 % Sat."] < 30 &
>                 x[,"Dia."] > 60 &
>                 (x[,"B635 Mean"] < (1.2 * x[,"B635 Median"]) &
>                  x[,"B532 Mean"] < (1.2 * x[,"B532 Median"])) &
>                 (x[,"Ratio of Medians (635/532)"] < (1.2 * x[,"Rgn
Ratio
>(635/532)"]) |
>                 x[,"Ratio of Medians (635/532)"] > (0.8 * x[,"Rgn
Ratio
>(635/532)"])) &
>                 x[,"Sum of Medians (635/532)"] > 1000
>                 )
>         ifnotflag <- x$Flags >= 0
>         return(as.numeric(ifnotflag & ifforeground))
>         }
>
>RG <- read.maimages(targets, source="genepix", columns=list(Rf="F635
>Median",
>         Gf="F532 Median", Rb="B635 Median", Gb="B532 Median"),
>         other.columns=list("B635 SD", "B532 SD", "F635 Median - B635",
>         "F532 Median - B532", "% > B635+2SD", "% > B532+2SD", "F635 %
Sat.",
>         "F532 % Sat.", "Dia.", "Ratio of Medians (635/532)", "Rgn
Ratio
>(635/532)",
>         "B635 Mean", "B635 Median", "B532 Mean", "B532 Median",
>         "Median of Ratios (635/532)", "Sum of Medians (635/532)"),
wt.fun =
>filter)

_______________________________________________
Bioconductor mailing list
Bioconductor at stat.math.ethz.ch
https://stat.ethz.ch/mailman/listinfo/bioconductor
Search the archives:
http://news.gmane.org/gmane.science.biology.informatics.conductor
=======================================================================
Attention: The information contained in this message and/or ...{{dropped}}