Run topic modeling with GSEA result
runLDA( SeuratObj, by = "GO", k = NULL, method = "VEM", SEED = 1234, plot = TRUE, BP.only = FALSE )
| SeuratObj | Object of class "Seurat" |
|---|---|
| by | using which GSEA result to run topic modeling, one of "GO", "KEGG", "Reactome", "MSigDb", "WikiPathways", "DO", "NCG", "DGN" |
| k | number of topics. Can be a specific k, if NULL, calculate best k automatically(time-consuming) |
| method | method used for fitting a LDA model; currently "VEM" or "Gibbs" are supported. |
| SEED | seed |
| plot | whether or not to plot metrics used to decide the number of topics |
| BP.only | use only pathways of Biological Process (BP) to run topic modeling. Take effect when by = "GO". |
a Seurat object with topic modeling result stored in the SeuratObj@misc$ldaOut
if (FALSE) { k <- 10 SeuratObj <- runLDA(SeuratObj, by = "GO", k = k, method = "VEM", SEED = 1234, plot = T) SeuratObj@misc$ldaOut # where the topic modeling result is stored }