Probabilistic quotient normalization — pqn

Apply probabilistic quotient normalization (PQN) to the peak table of a SummarizedExperiment or MetaboSet object. By default, reference is calculated from high-quality QC samples and the median of the reference is used for normalization. Check parameters for more options.

Usage

pqn_normalization(
  object,
  ref = c("qc", "all"),
  method = c("median", "mean"),
  all_features = FALSE,
  assay.type = NULL,
  name = NULL
)

Arguments

object: a SummarizedExperiment or MetaboSet object
ref: character, the type of reference samples to use for normalization.
method: character, the method to use for calculating the reference sample.
all_features: logical, should all features be used for calculating the reference sample?
assay.type: character, assay to be used in case of multiple assays
name: character, name of the resultant assay in case of multiple assays

Value

A SummarizedExperiment or MetaboSet object with altered feature abundances.

Examples

data(example_set)
pqn_set <- pqn_normalization(example_set)
#> INFO [2025-06-23 22:38:11] Starting PQN normalization
#> INFO [2025-06-23 22:38:11] Using median of qc samples as reference spectrum