plot_variance_explained: Plot the variance explained of sample metadata properties in...

In ftwkoopmans/msdap: Mass Spectrometry Downstream Analysis Pipeline

Plot the variance explained of sample metadata properties in the protein-intensity matrix

Description

precondition: the input dataset$peptides tibble must have a "intensity_all_group" column, e.g. obtained by first calling the filter_dataset() function with parameter all_group=TRUE (when using this function to analysis_quickstart() that is automatically taken care of)

Usage

plot_variance_explained(
  dataset,
  cols_metadata = NULL,
  rollup_algorithm = "maxlfq",
  quiet = TRUE
)

Arguments

dataset	your dataset. Prior to calling the this function, you must have applied "all_group" filtering and normalization using the filter_dataset() function (see example below)
cols_metadata	columns from 'dataset@samples“ to be used. Set to NA or NULL to automatically infer (default)
rollup_algorithm	algorithm for combining peptides to proteins as used in DEA algorithms that require a priori rollup from peptides to a protein-level abundance matrix before applying statistics (e.g. ebayes, deqms). Refer to rollup_pep2prot function documentation for available options and a brief description of each
quiet	boolean value, passed to variancePartition package

Details

1. performs peptide-to-protein rollup

2. impute missing values using the missForest R package (doi: 10.1093/bioinformatics/btr597)

3. call the variancePartition R package (Hoffman GE, Schadt EE, 2016, PMID:27884101)

note; this is quite slow even on small datasets

Examples

## Not run: 
  dataset = filter_dataset(
    dataset, filter_min_detect = 0, filter_min_quant = 3,
    by_group = FALSE, by_contrast = FALSE, all_group = TRUE,
    norm_algorithm = c("vwmb", "modebetween_protein")
  )
  tmp = plot_variance_explained(dataset, cols_metadata = NA)
  print(tmp$p_ve_violin)
  print(tmp$tbl_ve)

## End(Not run)

ftwkoopmans/msdap documentation built on March 5, 2025, 12:15 a.m.