Iterative consensus clustering

iter_consensus_clust(
  cl.list,
  co.ratio = NULL,
  cl.mat = NULL,
  norm.dat,
  select.cells = names(cl.list[[1]]),
  diff.th = 0.25,
  prefix = NULL,
  method = c("auto", "louvain", "ward.D"),
  verbose = FALSE,
  de.param = de.param,
  max.cl.size = 300,
  result = NULL,
  split.size = de.param$min.cells * 2,
  merge.type = c("undirectional", "directional")
)

Arguments

cl.list: The list of subsampled clustering results.
co.ratio: cell cell co-clustering matrix
norm.dat: The log2 transformed normalzied expression matrix
select.cells: Cells to be clustered
diff.th: The difference of co-clustering probablities for splitting a cluster.
prefix: Default NULL.
method: Clustering methods. Default "auto"
verbose: Default FALSE
de.param: Differentiall expressed genes criteria for merging clusters
result: Pre-computed clustering results used for further splitting. Default NULL.
merge.type: Determine if the DE gene score threshold should be applied to combined de.score, or de.score for up and down directions separately.

Value

A list with cluster membership, and top pairwise marker genes.