Runs RunSVD on the embedded Seurat object and
returns the updated MatisseObject. Performs Latent Semantic
Indexing (LSI) — the standard dimensionality-reduction step for ATAC-seq
data. Typically called after RunTFIDF.MatisseObject.
Usage
# S3 method for class 'MatisseObject'
RunSVD(
object,
assay = NULL,
features = NULL,
pca = FALSE,
n = 50,
reduction.key = ifelse(pca, "PCA_", "LSI_"),
scale.max = NULL,
verbose = TRUE,
...
)Arguments
- object
A
MatisseObject.- assay
Which assay to use. If NULL, use the default assay
- features
Which features to use. If NULL, use variable features
- pca
Run PCA. Setting this option to TRUE will perform implicit scaling and centering of the input matrix to enable memory-efficient computation of the principal components.
- n
Number of singular values to compute
- reduction.key
Key for dimension reduction object
- scale.max
Clipping value for cell embeddings. Default (NULL) is no clipping.
- verbose
Print messages
- ...
Additional arguments forwarded to
RunSVD(e.g.n,assay).