Skip to contents

plot PCA

Source: R/tidyMS_plotting.R
plot_pca.Rd

plot PCA

Usage

plot_pca(
  matrix,
  annotation,
  sample_name,
  factor_keys,
  PC = c(1, 2),
  add_txt = FALSE,
  nudge = 0.1
)

Arguments

matrix

numeric matrix — wide-format intensity data

annotation

data.frame — sample annotation

sample_name

character — sample name column

factor_keys

character vector — factor column names (first for color, second for shape)

PC

which principal components to plot

add_txt

show sample labels

nudge

label nudge distance

See also

Other plotting: ContrastsPlotter, INTERNAL_FUNCTIONS_BY_FAMILY, medpolish_estimate_df(), missigness_histogram(), missingness_per_condition(), missingness_per_condition_cumsum(), plot_estimate(), plot_heatmap(), plot_heatmap_cor(), plot_hierarchies_add_quantline(), plot_hierarchies_boxplot_df(), plot_hierarchies_line(), plot_hierarchies_line_df(), plot_intensity_distribution_violin(), plot_na_heatmap(), plot_raster(), plot_sample_correlation(), upset_interaction_missing_stats(), upset_missing_stats()

Examples


istar <- sim_lfq_data_protein_config(with_missing = TRUE, weight_missing = .8, Nprot = 3000)
#> creating sampleName from file_name column
#> completing cases
#> completing cases done
#> setup done
lfq <- LFQData$new(istar$data, istar$config)
wide <- lfq$to_wide(as.matrix = TRUE)
tmp <- plot_pca(wide$data, wide$annotation, lfq$sample_name(), lfq$factor_keys(),
  add_txt = TRUE, nudge = 0.01)
#> PCA: removed 2459 of 2986 features with missing values. For PCA with all features, impute first using impute_with_zcomp().
#> Joining with `by = join_by(sampleName)`
stopifnot("ggplot" %in% class(tmp))
tmp <- plot_pca(wide$data, wide$annotation, lfq$sample_name(), lfq$factor_keys())
#> PCA: removed 2459 of 2986 features with missing values. For PCA with all features, impute first using impute_with_zcomp().
#> Joining with `by = join_by(sampleName)`
stopifnot("ggplot" %in% class(tmp))