reselectMG: Reselect markers by thresholding
In CAMTHC: Convex Analysis of Mixtures for Tissue Heterogeneity Characterization

Description Usage Arguments Details Value Examples

This function generates a new list of markers based on initially detected markers by CAM and/or prior markers.

1	reselectMG(data, MGlist, fc.thres = "q0.5", err.thres = NULL)

`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 CAM-detected markers and/or prior markers for one subpopulation.
`fc.thres`	The lower threshold of fold change to select markers, an absolute value (e.g. 10) or a quantile value after 'q' (e.g. 'q0.5', the median of fold changes of one subpopulation's input markers). Each subpopulation can have its own threshold if a vector is provided. The default is 'q0.5'. If NULL, still use the input markers.
`err.thres`	The upper threshold of reconstruction error to select markers, an absolute value or a quantile value after 'q' (e.g. 'q0.5', the median of reconstruction errors of one subpopulation's input markers; 'q1.2', the maximum error times 1.2 ). Each subpopulation can have its own threshold if a vector is provided. The default is NULL, which means no such a threshold is applied.

Considering some meaningful markers may be mistakenly filtered by preprocessing and thus missed by CAM, this function use the input marker gene list to estimate proportions by AfromMarkers and then estimate expression levels. Next, a new list of markers are generated by fold change threshold and reconstruction error threshold.

The input marker genes could also be from other supervised detection and/or from literatures. This function reselects a list of marker genes based on the input.

A list of vectors, each of which contains new selected markers for one subpopulation.

#obtain data and run CAM
data(ratMix3)
data <- ratMix3$X
rCAM <- CAM(data, K = 3, dim.rdc= 3, thres.low = 0.30, thres.high = 0.95)
#obtain marker genes detected by CAM with a fixed K
MGlist <- MGsforA(rCAM, K = 3)
#Reselect markers from all genes
MGlist.re <- reselectMG(data, MGlist, fc.thres='q0.5')
MGlist.re <- reselectMG(data, MGlist, fc.thres='q0.5', err.thres='q0.95')