Fold-change scatterplot between two comparisons — get_foldchange

Plots log2 fold changes from two stored comparisons against each other, with points coloured by concordant/discordant regulation based on the cutoffs saved in the VISTA object.

Usage

get_foldchange_scatter(
  x,
  sample_comparisons,
  label_n = 0,
  alpha = 0.5,
  geometry = c("point", "hex"),
  method = c("pearson", "spearman"),
  colors = c(`Up/Up` = "#1b9e77", `Down/Down` = "#7570b3", `Up/Down` = "#d95f02",
    `Down/Up` = "#e7298a", Other = "grey70"),
  point_size = 1.5,
  label_size = 3,
  base_size = 12
)

Arguments

x: A VISTA object containing DE results.
sample_comparisons: Character vector of length 2 naming the comparisons.
label_n: Integer; number of most extreme points to label (by |log2FC1| + |log2FC2|).
alpha: Point transparency.
geometry: Geometry used for the data layer: "point" or "hex".
method: Correlation method for the subtitle; "pearson" or "spearman".
colors: Named vector of concordance colours.
point_size: Point size used when geometry = "point".
label_size: Text size for labeled genes.
base_size: Base theme size.

Value

A ggplot2 object.

Details

Points are coloured by concordance status using fixed colours:

Up/Up = #1b9e77
Down/Down = #7570b3
Up/Down = #d95f02
Down/Up = #e7298a
Other = grey70

Regulation is derived from the log2fc and pval cutoffs stored in cutoffs(x) (and p_value_type from the same list, defaulting to "padj").

Examples

if (FALSE) { # \dontrun{
data('count_data', package = 'VISTA')
data('sample_metadata', package = 'VISTA')
cell_levels <- unique(sample_metadata$cell)
if (length(cell_levels) >= 3) {
  v <- create_vista(count_data[1:150, ], sample_metadata, column_geneid = 'gene_id', group_column = 'cell',
                    group_numerator = cell_levels[2:3], group_denominator = rep(cell_levels[1], 2),
                    min_counts = 5, min_replicates = 1)
  comp_names <- names(comparisons(v))[1:2]
  p <- get_foldchange_scatter(v, sample_comparisons = comp_names)
  print(p)
}
} # }