[BioC] MANOR: information about id.rep
Pierre Neuvial
Pierre.Neuvial at curie.fr
Fri May 25 09:16:22 CEST 2007
Hello Michela,
Ballardini Michela wrote:
> ...
>
> I have two questions:
>
> 1. what’s na.rm?
- The parameter "na.rm" stands for "remove non-assigned values (NA)", it is
quite standard in R, consider the following example:
> x <- c(NA, 1, 2, 3)
> median(x)
[1] NA
> median(x, na.rm=TRUE)
[1] 2
>From the help on function median: "na.rm: a logical value indicating whether
'NA' values should be stripped before the computation proceeds."
In your case the parameter na.rm is passed to function 'FUN' when you call
norm.arrayCGH:
> arrayAB.norm <- norm.arrayCGH(arrayAB, var="LogRatio", FUN=median, na.rm=TRUE)
That is, the log-ratio of each of your 3588 clones will be calculated as the
median of the corresponding spots on the array, and NAs will be discarded
before the calculation.
> 2. where can I found the “single target value computed for all the
> replicates”? Is it arrayAB.norm$cloneValues$LogRatio?
Yes ! arrayAB.norm$cloneValues$LogRatio is a vector of length 3588 with
normalised log-ratios aggregated by clone. FYI,
arrayAB.norm$arrayValues$LogRatioNorm is a vector of length 21632 with
normalised log-ratios *before* aggregation by clone.
Hope this helps,
Pierre.
More information about the Bioconductor
mailing list