multiNMF: Run NMF on a list of Seurat objects

Description

Given a list of Seurat objects, run non-negative matrix factorization on each sample individually, over a range of target NMF components (k).

Usage

multiNMF(
  obj.list,
  assay = "RNA",
  slot = "data",
  k = 5:6,
  hvg = NULL,
  nfeatures = 2000,
  L1 = c(0, 0),
  min.exp = 0.01,
  max.exp = 3,
  center = FALSE,
  scale = FALSE,
  min.cells.per.sample = 10,
  hvg.blocklist = NULL,
  seed = 123
)

Value

Returns a list of NMF programs, one for each sample and for each value of 'k'. The format of each program in the list follosw the structure of nmf factorization models.

Arguments

obj.list: A list of Seurat objects
assay: Get data matrix from this assay
slot: Get data matrix from this slot (=layer)
k: Number of target components for NMF (can be a vector)
hvg: List of pre-calculated variable genes to subset the matrix. If hvg=NULL it calculates them automatically
nfeatures: Number of HVG, if calculate_hvg=TRUE
L1: L1 regularization term for NMF
min.exp: Minimum average log-expression value for retaining genes
max.exp: Maximum average log-expression value for retaining genes
center: Whether to center the data matrix
scale: Whether to scale the data matrix
min.cells.per.sample: Minimum numer of cells per sample (smaller samples will be ignored)
hvg.blocklist: Optionally takes a vector or list of vectors of gene names. These genes will be ignored for HVG detection. This is useful to mitigateeffect of genes associated with technical artifacts and batch effects (e.g. mitochondrial), and to exclude TCR and BCR adaptive immune(clone-specific) receptors. If set to `NULL` no genes will be excluded
seed: Random seed

Examples

Run this code

library(Seurat)
data(sampleObj)
geneNMF_programs <- multiNMF(list(sampleObj), k=5)

Run the code above in your browser using DataLab