Get the marker genes for each cluster

Source: R/MarkerGenes.R

Get the marker genes for each cluster and report their 1-normalized expression value.

GetMarkerTable(
  counts_data,
  cluster_labels,
  H = NULL,
  n_sorted = 50,
  gene_expression_threshold = NULL,
  use_H = FALSE,
  lognorm = FALSE
)

Arguments

counts_data

a matrix of expression values for each cell (rows) and gene (columns)
cluster_labels

a vector of cluster labels
H

a nonnegative matrix such that W = H*t(H), H_i,j is the cells_weight by which cell i belongs to the jth cluster
n_sorted

the number of markers to report in the n_sorted table
gene_expression_threshold

for n cells, for gene_expression_threshold = m, dont consider genes expressed in more than n-m cells or genes expressed in less than m cells
use_H

whether to use H as loading, default is false, instead using cumulative absolute difference to sort max-mean assigned labels
lognorm

perform log transform, normalization, and scaling on the data before the computing the markers

Value

a list containing

all

all the markers listed by gene index, cluster, and score

n_sorted

a subset of the sorted marker table with gene symbols