Select the number of clusters

Select the number of clusters by specified criteria.

selectModel(obj, criteria = 'MBIC', pen.const=1)
  # S3 method for drscObject
selectModel(obj, criteria = 'MBIC', pen.const=1)
  # S3 method for Seurat
selectModel(obj, criteria = 'MBIC', pen.const=1)

Arguments

obj: an object with class Seurat by DR.SC or class drscObject by DR.SC_fit.
criteria: a string, specify the criteria used for selecting the number of clusters, supporting "MBIC", "BIC" and "AIC".
pen.const: an optional positive value, the adjusted constant used in the MBIC criteria. It usually takes value between 0.1 to 1.

Value

For S3 method of Seurat, it return a revised "Seurat" object with updated Idents(seu), spatial.drsc.cluster in the metadata and DimReduc object named dr-sc in the slot reductions. For S3 method of drscObject, it returns a list with the following components:

bestK: the selected number of clusters.
cluster: inferred class labels
hZ: extracted latent features.
icMat: a numeric matrix including the criteria value for each number of clusters K.

References

Wei Liu, Xu Liao, Yi Yang, Huazhen Lin, Joe Yeong, Xiang Zhou, Xingjie Shi & Jin Liu (2022). Joint dimension reduction and clustering analysis of single-cell RNA-seq and spatial transcriptomics data, Nucleic Acids Research, gkac219.

Author

Wei Liu

Note

nothing

Examples

seu <- gendata_RNAExp(height=10, width=10,p=50, K=4)
library(Seurat)
seu <- NormalizeData(seu, verbose=FALSE)
# or choose 40 spatailly variable features using FindSVGs in DR.SC
seu <- FindSVGs(seu, nfeatures = 40, verbose=FALSE)
# users define the adjacency matrix
Adj_sp <- getAdj(seu, platform = 'ST')
#> Neighbors were identified for 100 out of 100 spots.
var.features <- seu@assays$RNA@var.features
X <- Matrix::t(seu[["RNA"]]@data[var.features,])
# maxIter = 2 is only used for illustration, and user can use default.
drscList <- DR.SC_fit(X,Adj_sp=Adj_sp ,K=4, maxIter=2, verbose=TRUE)
#> Fit DR-SC model...
#> -------------------Calculate inital values-------------
#> Using accurate PCA to obtain initial values
#> -------------------Finish computing inital values------------- 
#> -------------------Starting  ICM-EM algortihm-------------
#> iter = 2, loglik= -981.827503, dloglik=1.000000 
#> -------------------Complete!-------------
#> elasped time is :0.04
#> Finish DR-SC model fitting
drsc1 <- selectModel(drscList)
str(drsc1)
#> List of 4
#>  $ bestK  : Named num 4
#>   ..- attr(*, "names")= chr "K"
#>  $ cluster: num [1:100] 1 1 1 2 3 3 3 1 1 1 ...
#>  $ hZ     : num [1:100, 1:15] -7.94 -6.66 -9.91 14.41 11.98 ...
#>  $ icMat  : num [1, 1:2] 4 10653
#>   ..- attr(*, "dimnames")=List of 2
#>   .. ..$ : chr "K"
#>   .. ..$ : chr [1:2] "K" "IC"