sig.mask {OLIN} | R Documentation |
This function sets data to NA if the corresponding spots have significantly biased neighbourhoods on the intensity scale or on the spatial dimensions of the array.
sig.mask(object,Sp,Sn,thrp,thrn)
object |
object of class marrayRaw or marrrayNorm |
Sp |
list of vectors of false discovery rate or p-values for positive deviation of median/mean of M
as produced by fdr.int2, p.int2, fdr.spatial2 or p.spatial2 . |
Sn |
list vector of false discovery rate or p-values for negative deviation of median/mean of M
as produced by fdr.int2, p.int2, fdr.spatial2 or p.spatial2 . |
thrp |
vector of thresholds for significance of positive deviation (Sp ) |
thrn |
vector of thresholds for significance of negative deviation (Sn ) |
This function can be used for the masking of data that has been decided to be unrelaible
after the application of significance test for intenstiy- and location dependent dye bias (e.g. p.int2, fdr.int2,
p.spatial2, fdr.spatial2
).
Matthias E. Futschik (http://itb.biologie.hu-berlin.de/~futschik)
sigint.plot
, fdr.int
, p.int
, sigxy.plot
,
fdr.spatial
, p.spatial
# To run these commands, delete comment sign (#) ! # # LOADING DATA # data(sw) # # MASKING REGIONS WITH SPATIAL DYE BIAS # # CALCULATION OF SIGNIFICANCE OF SPOT NEIGHBOURHOODS # For this example, N was chosen rather small. For "real" analysis, it should be larger. # FDR <- fdr.spatial2(sw,delta=2,N=10,av="median",edgeNA=FALSE) # # VISUALISATION # sigxy.plot2(sw[,1],FDR$FDRp[[1]],FDR$FDRn[[1]],color.lim=c(-5,5),main="FDR") # # MASKING SIGNIFICANT NEIGHBOURHOODS # thresp <- c(0.01,0.01,0.01,0.01) # thresn <- c(0.01,0.01,0.01,0.01) # sw.masked <- sig.mask(sw,Sp=FDR$FDRp,Sn=FDR$FDRn,thrp=thresp,thrn=thresn) # mxy.plot(sw.masked[,4]) # plot masked data for array 4