TileCalcWinTrim {TileShuffle} | R Documentation |
Calculates arithmetic average trimmed by the minimal and maximal value on probe scores with window indices.
TileCalcWinTrim(int, win.idx)
int |
A numeric vector containing one or more probe
scores (i.e. intensities or log-fold changes). |
win.idx |
A list with the maximal number of probes in a window
as its length. In entry i , it stores only start positions of
windows containing at least (i+1) probes. |
Calculates arithmetic average trimmed by the minimal and maximal value on
probe scores with window indices, a list
storing window start
positions. More precisely, win.idx
has the maximal number of probes
in a window as length and in its i
th entry, it stores only start
positions of windows containing at least (i+1)
probes. In case of
short but numerous windows, the window scores can be calculated very
efficiently by use of optimized vector
operations. Otherwise, please
consider using TileCalcWinTrim2
instead. Note that the
definition of the common'trimmed' mean differs from the common one with given
percentile ranges. Moreover, the resulting scores with trimmed mean may only
differ from the mean only in case of windows that contain more than two
probes.
A numeric
vector
containing the arithmetic mean trimmed
by the minimal and maximal value of the window scores for each
possible start position. Note that windows of length zero are not
supported and hence treated as windows of length one. Moreover, probe
scores with the value NA
are excluded from the calculation.