create_som: Create Self-Organizing Map

In biosurf/cyCombine: Robust integration of single-cell cytometry datasets within and across technologies

Create Self-Organizing Map

Description

The function uses the kohonen package to create a Self-Organizing Map. It is used to segregate the cells for the batch correction to make the correction less affected by samples with high abundances of a particular cell type.

Usage

create_som(df, markers = NULL, seed = 473, rlen = 10, xdim = 8, ydim = 8)

Arguments

df	tibble with expression values
markers	Markers to normalize. If NULL, markers will be found using the get_markers function.
seed	The seed to use when creating the SOM.
rlen	Number of times the data is presented to the SOM network
xdim	The x-dimension size of the SOM.
ydim	The y-dimension size of the SOM.

Value

A vector of clustering labels

See Also

Other batch: batch_correct(), correct_data(), normalize(), quantile_norm()

Examples

## Not run: 
labels <- uncorrected %>%
  create_som()
  
## End(Not run)

biosurf/cyCombine documentation built on May 23, 2024, 4:07 a.m.