NCFM_fast: Cell-feature coembedding for SRT data
In ProFAST: Probabilistic Factor Analysis for Spatially-Aware Dimension Reduction

NCFM_fast

R Documentation

Cell-feature coembedding for SRT data

Description

Run cell-feature coembedding for SRT data based on FAST model.

Usage

NCFM_fast(
  object,
  Adj_sp,
  assay = NULL,
  slot = "data",
  nfeatures = 2000,
  q = 10,
  reduction.name = "fast",
  var.features = NULL,
  ...
)

Arguments

`object`	a Seurat object.
`Adj_sp`	a sparse matrix, specify the adjacency matrix among spots.
`assay`	an optional string, the name of assay used.
`slot`	an optional string, the name of slot used.
`nfeatures`	an optional postive integer, the number of features to select as top variable features. Default is 2000.
`q`	an optional positive integer, specify the dimension of low dimensional embeddings to compute and store. Default is 10.
`reduction.name`	an optional string, dimensional reduction name, 'fast' by default.
`var.features`	an optional string vector, specify the variable features, used to calculate cell embedding.
`...`	Other argument passed to the `FAST_run`.

Examples

data(CosMx_subset)
pos <- as.matrix(CosMx_subset@meta.data[,c("x", "y")])
Adj_sp <- AddAdj(pos)
# Here, we set maxIter = 3 for fast computation and demonstration.
CosMx_subset <- NCFM_fast(CosMx_subset, Adj_sp = Adj_sp, maxIter=3)

ProFAST documentation built on May 29, 2024, 7:15 a.m.