Dear list, I discovered an oddity in snprma (oligo, svn revision 26822). It appears that ranks are used to extract elements from vectors in for (i in 1:ncol(tmpExprs)) tmpExprs[, i] <- reference[rank(tmpExprs[, i])] This might break iff there are two values that are exactly equal as can be seen from bar <- 2 * c(2, 4, 4, 5, 1) rank(bar) foo <- 1:5 foo[bar] A solution would be to specify an alternative ties.method such as rank(bar, ties.method = "first") Any thoughts? Regards, Tobias -- Tobias Verbeke - Consultant Business & Decision Benelux Rue de la r?volution 8 1000 Brussels - BELGIUM +32 499 36 33 15 tobias.verbeke at businessdecision.com

Hi Tobias, sorry for the late reply. I've just committed a change which uses preprocessCore:::normalize.quantiles.use.target instead of the code mentioned below. thanks, b On Sep 2, 2007, at 3:18 AM, Tobias Verbeke wrote: > Dear list, > > I discovered an oddity in snprma (oligo, > svn revision 26822). It appears that ranks > are used to extract elements from vectors in > > for (i in 1:ncol(tmpExprs)) > tmpExprs[, i] <- reference[rank(tmpExprs[, i])] > > > This might break iff there are two values that > are exactly equal as can be seen from > > bar <- 2 * c(2, 4, 4, 5, 1) > rank(bar) > foo <- 1:5 > foo[bar] > > A solution would be to specify an alternative > ties.method such as > > rank(bar, ties.method = "first") > > Any thoughts? > > Regards, > Tobias > > -- > > Tobias Verbeke - Consultant > Business & Decision Benelux > Rue de la r?volution 8 > 1000 Brussels - BELGIUM > > +32 499 36 33 15 > tobias.verbeke at businessdecision.com > > _______________________________________________ > 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