Skip to contents

Volcano plot of the genes

Source: R/plot_volcano.R
plot_volcano.Rd

Shows every tested gene with its effect (the mean log2 fold change of its barcodes, x axis) and its significance (-log10 FDR, y axis), so that the hits stand out from the other genes. It works with the table of a single method (find_roast_hit, find_camera_hit, find_gene_zscore_hit) or with the combined table of summarise_hits.

Usage

plot_volcano(
  table,
  statistics = NULL,
  fdr = 0.05,
  fdr_column = NULL,
  labels = 10,
  genes = NULL,
  colors = c(Hit = "#C9836A", `Not a hit` = "#93B5A3"),
  label_color = "black",
  alpha = 0.7,
  point_size = 1.8,
  label_size = 3.5
)

Arguments

table

A table with one row for each gene and the columns Gene, the FDR (see fdr_column) and Log2FC (unless statistics is given). Computed with fdr = 1, so that it contains all the tested genes.

statistics

Optional. The barcode statistics of the comparison (compute_barcode_statistics), with the columns Gene and Log2FC: the mean log2 fold change of the barcodes of each gene is used as its effect. Needed when table has no Log2FC column.

fdr

The FDR threshold that defines the hits (default 0.05), drawn as a dashed line.

fdr_column

The column with the FDR. By default "MaxFDR" when the table has it, otherwise "FDR".

labels

The number of hits (the most significant) labelled with their name (default 10, 0 for none).

genes

Optional names of genes to highlight with a circle and a label, whether they are hits or not.

colors

The colours of the hits and of the other genes.

label_color

The colour of the labels and of the circles of the highlighted genes (default "black").

alpha

The transparency of the points (default 0.7).

point_size

The size of the points (default 1.8).

label_size

The size of the labels (default 3.5).

Value

A ggplot object.

Details

The FDR is taken from fdr_column: by default MaxFDR when the table has it (summarise_hits: the largest FDR of the methods, so a gene is above the dashed line only when every method finds it) and FDR otherwise. The effect is the Log2FC column of the table (the median log2 fold change of the barcodes for FRY, ROAST and CAMERA, the mean for the Z-score test). The table of summarise_hits has none: give then the barcode statistics with statistics, and the mean log2 fold change of the barcodes of each gene is used (also with the other tables, when statistics is given).

Genes without an FDR (for example genes not tested by every method) are not shown. FDRs of 0 are drawn at the smallest positive FDR of the table. No theme is applied; choose the appearance with ggplot2::theme_set().

See also

summarise_hits, plot_common_hit, plot_barcode_hit

Examples

data(count_table)
data(annotation_table)
# The unmapped reads "*" appear several times: remove them
counts <- count_table[count_table$Barcode != "*", ]
groups <- factor(c(
    "T1/T2", "T1/T2", "Time3_TRT", "Time3_TRT", "Time3_TRT",
    "Time3_CTRL", "Time3_CTRL", "Time3_CTRL", "Time4_TRT", "Time4_TRT",
    "Time4_TRT", "Time4_CTRL", "Time4_CTRL", "Time4_CTRL"
))
object <- create_screenr_object(
    table = counts, annotation = annotation_table, groups = groups
)
#> The Day and the Treatment of each sample are taken from the sample names (Day_Treatment_Replicate). For 2 sample(s) (Time_1, Time_2) they cannot be separated and are both set to the first part of the name. If this is not intended, give them with the sample_table argument. Check them with get_sample_table().
object <- normalize_data(object)
design <- create_design_matrix(object)
contrast <- create_contrast(design,
    treatment = "Time4_TRT", control = "Time4_CTRL"
)

# One method
camera <- find_camera_hit(object, design, contrast, fdr = 1)
#> Warning: 3rows with all zero counts
#> Warning: Zero sample variances detected, have been offset away from zero
plot_volcano(camera, fdr = 0.25)


# Several methods: a gene is a hit when every method finds it
statistics <- compute_barcode_statistics(object, design, contrast)
zscore <- find_gene_zscore_hit(statistics, reference = "empirical",
    fdr = 1
)
hits <- summarise_hits(list(camera = camera, zscore = zscore), fdr = 0.25)
plot_volcano(hits, statistics = statistics, fdr = 0.25)