Partial Gene Expression Deconvolution: Marker Mean Expression Profile

Description

The algorithm ‘meanProfile’ uses a set of known marker genes for each cell type to compute a mean expression profile within each cell type separately.

This is in essence the preliminary step proposed by Kuhn et al. (2011) in order to compute a proxy for the actual cell proportions.

The average profiles are expected to be correlate well with the actual proportions, provided that the individual markers gene expression profiles are not too noisy and that the markers are indeed markers.

Usage

gedAlgorithm.meanProfile(y, x, scale = FALSE, ...)

Arguments

y
target expression matrix
x
specification of which feature to use as markers, as a list, typically a MarkerList object.
scale
logical that indicates if the proportion estimates should be scale to sum-up to one before returning them.
...
extra arguments not used.

Details

Important: this method does not compute cell-specific differential expression as described in Kuhn et al. (2011), but only the cell proportion proxy. Hence, the result only contains estimated proportions (accessible via coef) and an empty basis signature matrix.

References

Kuhn A, Thu D, Waldvogel HJ, Faull RLM and Luthi-Carter R (2011). "Population-specific expression analysis (PSEA) reveals molecular changes in diseased brain." _Nature methods_, *8*(11), pp. 945-7. ISSN 1548-7105, , .

Examples


# random data with markers
x <- rmix(3)
m <- getMarkers(x)

# compute proxy proportions
res <- ged(x, m, method='meanProfile')
# no cell-specific signatures
dim(res)
## [1]  0 20  3

#NB: estimates are not scaled to sum up to one
profplot(x, res)

plot of chunk unnamed-chunk-1

profplot(x, res, scale=TRUE)

plot of chunk unnamed-chunk-1