Clusters_denovo: Assign new observations to the clusters inside a MOBSTER fit.

In caravagnalab/mobster: MOBSTER - Model-based tumour subclonal deconvolution

Assign new observations to the clusters inside a MOBSTER fit.

Description

For a new set of observations in the MOBSTER input format, the parameters and clusters of a MOBSTER fit are used to determine the assignments of the new points. MOBSTER density function is used to determine which cluster has the highest density for each observation, and the cluster label is computed accordingly.

Usage

Clusters_denovo(x, y)

Arguments

x	A MOBSTER fit.
y	A MOBSTER input dataset, which has to have a VAF numeric column without NAs.

Value

The data in y is augmented with a colum per mixture component reporting the corresponding density value. A final colum 'cluster' is also added reporting the component name for the hard clustering assignment of the point.

Examples

library(ggplot2)
data('fit_example', package = 'mobster')

# Generate some randome numbers and assign them to the most likely mixture component
new_assignments = Clusters_denovo(fit_example$best, data.frame(VAF = runif(1000))) 
print(new_assignments)

# Plot a histogram coloured according to the clusters
ggplot(new_assignments, aes(VAF, fill = cluster)) + geom_histogram(binwidth = 0.01)

caravagnalab/mobster documentation built on March 25, 2023, 3:40 p.m.