gapFilter-methods {xps} | R Documentation |
This method initializes the Gap Filter.
The gapFilter
looks for genes that might usefully discriminate between two groups.
To do this we look for a gap in the ordered expression values. The gap should come in the
central portion, thus a parameter window
is defined to exclude jumps in the initial
window
values and the final window
values.
The Gap Filter flags all rows with: flag = ((gap[i+1] - gap[i])/mean >= cutoff)
gapFilter(object)
gapFilter(object, value)<-
object |
object of class PreFilter . |
value |
numeric vector c(cutoff, window, trim, epsilon) . |
The method gapFilter
initializes the following parameters:
cutoff : | the cutoff level for the filter. | |
window : | trim value for the ordered expression levels (default is window=0.05 ). |
|
trim : | the trim value for trimmed mean (default is trim=0 ). |
|
epsilon : | value to replace mean (default is epsilon=0.01 ): |
|
epsilon > 0 : replace mean=0 with epsilon. |
||
epsilon = 0 : always set mean=1. |
Note, that for epsilon = 0
the filter flags all rows with: (gap[i+1] - gap[i]) >= cutoff
An initialized PreFilter
object.
Christian Stratowa
prefltr <- PreFilter() gapFilter(prefltr) <- c(0.3,0.05,0.0,0.01) str(prefltr)