AfromMarkers: Proportion matrix estimation from marker genes
In debCAM: Deconvolution by Convex Analysis of Mixtures

Description Usage Arguments Details Value Examples

This function estimates proportion matrix (A matrix) from observed mixture expression data based on marker genes.

1	AfromMarkers(data, MGlist, scaleRecover = TRUE)

`data`	A data set that will be internally coerced into a matrix. Each row is a gene and each column is a sample. data should be in non-log linear space with non-negative numerical values (i.e. >= 0). Missing values are not supported. All-zero rows will be removed internally.
`MGlist`	A list of vectors, each of which contains known markers and/or CAM-detected markers for one subpopulation.
`scaleRecover`	If TRUE, scale ambiguity of each column vector in A matrix is removed based on sum-to-one constraint on each row.

With the expression levels of subpopulation-specific marker genes, the relative proportions of constituent subpopulations are estimated by spatial median using l1median. Marker genes could be from unsupervised/supervised detection or from literatures. Scale ambiguity is optionally removed based on sum-to-one constraint of rows.

Return the estimated proportion matrix (A matrix).

#obtain data and marker genes
data(ratMix3)
S <- ratMix3$S
pMGstat <- MGstatistic(S, c("Liver","Brain","Lung"))
pMGlist.FC <- lapply(c("Liver","Brain","Lung"), function(x)
    rownames(pMGstat)[pMGstat$idx == x & pMGstat$OVE.FC > 10])

#estimate A matrix from markers
Aest <- AfromMarkers(ratMix3$X, pMGlist.FC)