embedCells: Generate cell-state embedding

In wschen/phemd: Phenotypic EMD for comparison of single-cell samples

Generate cell-state embedding

Description

Takes as input a Phemd object with aggregated data and returns updated object containing cell-state embedding

Usage

embedCells(
  obj,
  cell_model = c("monocle2", "seurat", "phate"),
  data_model = "negbinomial_sz",
  phate_ncluster = 8,
  phate_cluster_seed = NULL,
  ...
)

Arguments

obj	'Phemd' object containing aggregated data
cell_model	Method to use to generate cell-state embedding. Currently supports "phate" and "monocle2". If using the Seurat to model the cell-state space, please identify cell subtypes as outlined in the Seurat software package and then use the bindSeuratObj function.
data_model	Only relevant if cell_model = "monocle2". One of the following: 'negbinomial_sz', 'negbinomial', 'tobit', 'uninormal', 'gaussianff'. See "Family Function" table at the following link for more details on selecting the proper one. http://cole-trapnell-lab.github.io/monocle-release/docs/#getting-started-with-monocle
phate_ncluster	Only relevant if cell_model = "phate". Number of cell state clusters to return when using PHATE
phate_cluster_seed	Only relevant if cell_model = "phate". Seed to use when performing cell state clustering (optional)
...	Additional parameters to be passed to reduceDimension function for Monocle or phate function for PHATE

Details

aggregateSamples needs to be called before running this function.

Value

Same as input 'Phemd' object containing additional cell-state embedding object

Examples

my_phemdObj <- createDataObj(all_expn_data, all_genes, as.character(snames_data))
my_phemdObj_lg <- removeTinySamples(my_phemdObj, 10)
my_phemdObj_lg <- aggregateSamples(my_phemdObj_lg, max_cells=1000)
my_phemdObj_lg <- embedCells(my_phemdObj_lg, cell_model='monocle2', data_model = 'gaussianff', sigma=0.02, maxIter=2)

wschen/phemd documentation built on Oct. 1, 2024, 4:45 a.m.