Description Usage Arguments Details Value Note References See Also Examples
Calculate the bin-wise averages of the correlations between rows of a matrix. This may be useful for visualizing gross trends in the correlation matrix.
1 2 3 |
x |
A matrix, or an exprSet. |
n |
Number of bins. |
use, method |
Passed directly to |
FUN |
Function to summarize values in each bin. |
... |
Additional argments are passed to |
The rows of the input matrix x
are split into n
bins. For each possible pair of bins, all possible correlation coefficients between rows in the two bins are calculated. The median (or other FUN
) correlation for each pair of bins is returned as a matrix.
For the diagonals of the returned matrix, which correspond to the correlation between a bin and itself, only the correlations between non-identical probe sets are included.
A list with the following components:
x, y |
Numeric vectors of length |
z |
|
count |
|
call |
Character; the function call. |
The results will be more informative if the input matrix x
is sorted by some criterion first. For example, try sorting by (row-wise) mean, variance, probe sequence composition...
Eklund, A.C. and Szallasi, Z. (2008) Correction of technical bias in clinical microarray data improves concordance with known biological information. Genome Biology, 9:R26.
colorgram
in the in the squash package for visualization
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 | library(affydata)
data(Dilution)
d <- exprs(rma(Dilution))
## Sort the rows (probe sets) by mean expression, and calculate correlation matrix
dsort <- d[order(rowMeans(d)),]
cm <- calcCM(dsort)
## Visualize the correlation matrix
image(cm, col = rainbow(20),
xlab = 'Expression rank', ylab = 'Expression rank')
## Maybe use the "squash" package for more control of the plot
## Not run:
par(mar = c(7, 4, 4, 1) + 0.1)
map <- makecmap(cm$z, colFn = blueorange,
symm = TRUE, n = 20)
colorgram(cm, map = map,
xlab = 'Expression rank',
ylab = 'Expression rank',
zlab = 'Median Pearson correlation')
## End(Not run)
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.