CellMix 1.6

Partial Gene Expression Deconvolution by Least-Square

Description

.gedLSfit provides access to partial deconvolution methods that are based on least-squares fits.

.nn_lsfit implements a standard least-square fit with various different procedures to enforce nonnegative coefficients. In particular, it implements the iterative procedure described in Abbas et al. (2009).

Usage

.gedLSfit(X, seed, rescale = TRUE, fit = c("ls", "nnls"), ...)

.nn_lsfit(x, y, nneg = c("iterate", "pmax", "none"), ...)

Arguments

rescale
logical used when estimating proportions from signatures, that indicates if the esti,ated coefficients should be scaled to sum up to one (TRUE) or left as estimated by the linear regression (FALSE). This scaling is performed after the coefficients have been forced to be nonnegative.
fit
least-square fitting method: ls uses lm, nnls uses fcnnls.
...
extra arguments passed to fitting the methods .nn_lsfit or .fcnnls.
x
matrix of known cell-specific profiles (i.e. cell signatures), with features in rows and cell type in columns.
y
matrix of observed mixed expression data, with features in rows and samples in columns. The number of samples must be greater -- or equal -- than the number of cell types.
nneg
specification of the method used to enforce the nonnegativity of the estimated proportions. Accepted values are:
  1. 'iterate': applies the procedure described in Abbas2009. For each sample separately, a sequence of least-square fits are performed, starting with all cell types, and where the cell type corresponding to the lowest negative fitted coefficient is excluded from the next fit, and its associated final proportion set to zero. This iterative process stops when all coefficients are nonnegative.
  2. 'pmax': single least-square fit, where all negative estimated proportions are set to zero.
  3. NA or 'none': single least-square fit where the estimated proportions are returned unconstrained.
X
target matrix
seed
initial NMF model as an NMF object.

Value

an NMF object.

References

Abbas AR, Wolslegel K, Seshasayee D, Modrusan Z and Clark HF (2009). "Deconvolution of blood microarray data identifies cellular activation patterns in systemic lupus erythematosus." _PloS one_, *4*(7), pp. e6098. ISSN 1932-6203, , .