find_markers {scGPS}R Documentation

find marker genes

Description

Find DE genes from comparing one clust vs remaining

Usage

find_markers(expression_matrix = NULL, cluster = NULL,
  selected_cluster = NULL, fitType = "local",
  dispersion_method = "per-condition", sharing_Mode = "maximum")

Arguments

expression_matrix

is a normalised expression matrix.

cluster

corresponding cluster information in the expression_matrix by running CORE clustering or using other methods.

selected_cluster

a vector of unique cluster ids to calculate

fitType

string specifying 'local' or 'parametric' for DEseq dispersion estimation

dispersion_method

one of the options c( 'pooled', 'pooled-CR', per-condition', 'blind' )

sharing_Mode

one of the options c("maximum", "fit-only", "gene-est-only")

Value

a list containing sorted DESeq analysis results

Author(s)

Quan Nguyen, 2017-11-25

Examples

day2 <- day_2_cardio_cell_sample
mixedpop1 <-new_scGPS_object(ExpressionMatrix = day2$dat2_counts, 
    GeneMetadata = day2$dat2geneInfo, CellMetadata = day2$dat2_clusters)
# depending on the data, the DESeq::estimateDispersions function requires
# suitable fitType
# and dispersion_method options
DEgenes <- find_markers(expression_matrix=assay(mixedpop1),
                        cluster = colData(mixedpop1)[,1],
                        selected_cluster=c(1), #can also run for more
                        #than one clusters, e.g.selected_cluster = c(1,2)
                        fitType = "parametric", 
                        dispersion_method = "blind",
                        sharing_Mode="fit-only"
                        )
names(DEgenes)

[Package scGPS version 1.4.0 Index]