InferDMAPandRoot: Infers Diffusion Map, Root-state and root-cell

In aet21/SCENT: Single Cell Entropy

Infers Diffusion Map, Root-state and root-cell

Description

This function constructs the diffusion map, Markov transition matrix and uses the provided potency estimates to infer the root-state and root-cell.

Usage

InferDMAPandRoot(
  pot.v,
  exp.m,
  avth=1,
  sdth=0.25,
  kDMAP = 30,
  pctop = 0.05
)

Arguments

pot.v	A vector of potency estimates for all cells, e.g. a vector of SR or CCAT values.
exp.m	The normalized and log-transformed scRNA-Seq data matrix
avth	The threshold on the average expression to use for selecting genes in the diffusion map. Default values is 1.
sdth	The threshold on the standard deviation to use for selecting genes in the diffusion map. Default value is 0.25.
kDMAP	An integer specifying the input argument to the diffusion map, representing the number of nearest neighbors to consider. By default this value is 30.
pctop	The proportion of top-ranked cells (with cells ranked by SR or CCAT) to consider when inferring teh root-state. By default this values is 0.05.

Value

A list with the following elements:

dmap	The diffusion map object, as given by the 'destiny' package.
dc	The inferred diffusion map component matrix, with rows labeling cells and columns labeling ranked diffusion components.
transM	The Markov transition matrix associated with the inferred diffusion map. Rows and columns label cells.
root	The column index of the root-cell.
color	A color vector for the cells stratified according to potency

References

Teschendorff AE, Tariq Enver. Single-cell entropy for accurate estimation of differentiation potency from a cell’s transcriptome. Nature communications 8 (2017): 15599. doi: 10.1038/ncomms15599.

aet21/SCENT documentation built on Aug. 1, 2022, 12:05 p.m.