PlotPCA.overview() modifies original mSetObj. PlotPCA.overview() and EigenMS() truncate negative values.

[maciejrosolowski]

1. The following line in the function PlotPCA.overview modifies the original mSetObj:
   mSetObj$dataSet$adjusted.mat <- table;
   It should probably be removed. In addition, the next line:
   pca <- prcomp(mSetObj$dataSet$adjusted.mat, center=T, scale=T);
   should probably be replaced with:
   pca <- prcomp(table, center=T, scale=T);
   since "table" contains the values to be used in the PCA.

2. The following line in the function PlotPCA.overview:
   table[table < 10^-20] <- 0;
   truncates negative values at 0. Negative values can occur, if the normalized data were log-transformed. The above line could probably be removed or replaced with:
   table[abs(table) < 10^-20] <- 0;

3. The following line in the function EigenMS:
   data.table[data.table<0]<-0
   truncates negative values at 0. It should probably be removed.

Or there is a reason why the functions PlotPCA.overview and EigenMS set negative values to 0?