estimate_projection_failure: Estimate likelihood of projection failure

In quadbiolab/simspec: Calculate data representation of Reference/Cluster Similarity Spectrum

Estimate likelihood of projection failure

Description

Calculate the normalized distances of each query sample to the projected reference subset, and use a Gaussian mixture model to estimate the likelihood of projection failure, which is featured by large distance to the reference

Usage

estimate_projection_failure(
  data_ref,
  data_query,
  k = 50,
  seed = NULL,
  do_plot = FALSE,
  verbose = TRUE
)

Arguments

data_ref	Reference data matrix
data_query	Query data matrix
k	Number of neighbors
seed	The seed value for the Gaussian mixture model estimate
do_plot	Whether to plot the fitted mixture model (as density plot)
verbose	Whether to output the verbose message

Value

A data frame with normalized projection distance and estimated projection failure likelihood

quadbiolab/simspec documentation built on March 8, 2024, 11:59 p.m.