Classify: (Re)classify after running 'CHETAHclassifier' using a...

In jdekanter/CHETAH: Fast and accurate scRNA-seq cell type identification

Description

(Re)classify after running CHETAHclassifier using a confidence threshold
NOTE: In case of bulk reference profiles: only the correlations will be used, as the data does not allow for profile or confidence scores to be calculated.

Usage

Classify(input, thresh = 0.1, return_clas = FALSE)

Arguments

input	a SingleCellExperiment on which CHETAHclassifier has been run
thresh	a confidence threshold between -0 and 2. Selecting 0 will classify all cells, whereas 2 will result i n (almost) no cells to be classified. recommended: between 0.1 (fairly confident) and 1 (very confident)
return_clas	Instead of returning the SingleCellExperiment, only return the classification vector

Value

a charachter vector of the cell types with the names of the cells

Examples

## Classify all cells
input_mel <- Classify(input_mel, 0)

## Classify only cells with a very high confidence
input_mel <- Classify(input_mel, 1)

## Back to the default
input_mel <- Classify(input_mel)

## Return only the classification vector
input_mel <- Classify(input_mel, 1, return_clas = TRUE)

jdekanter/CHETAH documentation built on Sept. 20, 2019, 3:31 a.m.