scVIIntegration: scVI Integration

In satijalab/seurat-wrappers: Community-Provided Methods and Extensions for the Seurat Object

scVI Integration

Description

scVI Integration

Usage

scVIIntegration(
  object,
  features = NULL,
  layers = "counts",
  conda_env = NULL,
  new.reduction = "integrated.dr",
  ndims = 30,
  nlayers = 2,
  gene_likelihood = "nb",
  max_epochs = NULL,
  ...
)

Arguments

object	A StdAssay or STDAssay instance containing merged data
features	Features to integrate
layers	Layers to integrate
conda_env	conda environment to run scVI
new.reduction	Name under which to store resulting DimReduc object
ndims	Dimensionality of the latent space
nlayers	Number of hidden layers used for encoder and decoder NNs
gene_likelihood	Distribution to use for modelling expression data: "zinb", "nb", "poisson"
max_epochs	Number of passes through the dataset taken while training the model
...	Unused - currently just capturing parameters passed in from Seurat::IntegrateLayers intended for other integration methods

Value

A single-element named list DimReduc elements containing the integrated data

Note

This function requires the scvi-tools package to be installed

See Also

scVI

Examples

## Not run: 
# Preprocessing
obj <- SeuratData::LoadData("pbmcsca")
obj[["RNA"]] <- split(obj[["RNA"]], f = obj$Method)
obj <- NormalizeData(obj)
obj <- FindVariableFeatures(obj)
obj <- ScaleData(obj)
obj <- RunPCA(obj)

# After preprocessing, we integrate layers, specifying a conda environment
obj <- IntegrateLayers(
  object = obj,
  method = scVIIntegration,
  new.reduction = "integrated.scvi",
  conda_env = "../miniconda3/envs/scvi-env",
  verbose = FALSE
)

# Alternatively, we can integrate SCTransformed data
obj <- SCTransform(object = obj)
obj <- IntegrateLayers(
  object = obj, method = scVIIntegration,
  orig.reduction = "pca", new.reduction = "integrated.scvi",
  assay = "SCT", conda_env = "../miniconda3/envs/scvi-env", verbose = FALSE
)

## End(Not run)

satijalab/seurat-wrappers documentation built on April 10, 2024, 3:25 p.m.