plot_lollipop {goat}

Lollipop chart or barplot visualization of geneset enrichment testing results

Description

Lollipop chart or barplot visualization of geneset enrichment testing results

Usage

plot_lollipop(
  x,
  output_dir,
  only_reduced = FALSE,
  plot_type = "lollipop",
  show_pvalue = FALSE,
  score_xaxis = "minlogp",
  score_color = ifelse(is.data.frame(x) && "score_type" %in% colnames(x) &&
    is.character(x$score_type) && any(x$score_type %in% c("effectsize_up",
    "effectsize_down")), "updown", "minlogp"),
  score_color_limits = "source",
  score_color_updown = c("#E57373", "#5E7ABC"),
  max_ngenes = NA,
  topn = NA,
  padj_cutoff = NA
)

Arguments

x

results from function test_genesets

output_dir

set to NA to directly show the figures instead of writing them to file. Otherwise, this is the full path to the directory where the downloaded files should be stored. Directory is created if it does not exist. e.g. output_dir="~/data" on unix systems, output_dir="C:/data" on Windows, or set to output_dir=getwd() to write output to the current working directory

only_reduced

only show the reduced/summarized set of significant genesets. This requires that you first applied the reduce_genesets function

plot_type

Options: "barplot", "lollipop" (default)

show_pvalue

boolean parameter that indicates whether adjusted p-values should be shown next to each data point

score_xaxis

type of score to show on the x-axis. Options: "minlogp" to show -log10(adjusted pvalue), which is default. Use "oddsratio" to show the enrichment of significant genes. For further details on this score and its computation, see the score_geneset_oddsratio function documentation. Basically, the genesets in this plot are sorted by their proportion of foreground/significant genes (and this ratio is standardized against the overall ratio of significant genes as to make this statistic comparable across analyses/datasets).

score_color

analogous to score_xaxis, here you can specify the data used for color-coding the plot. Options: "minlogp", "oddsratio", "updown". The former 2 options are the same as for score_xaxis, the latter enables color-coding by up-/down-regulation as encoded in the "score_type" column. Note that this only works when geneset testing based on "effectsize" was performed and thus the "score_type" column has values "effectsize_up" or "effectsize_down" (encoding directionality). Rows with other values are assumed NA and colored as grey.

score_color_limits

defines the limits for the color scales. options; score_color_limits = "source" use the range of values per 'source' to compute colors scales (default). Set to "overall" in order to have a unified color scale across 'source' (e.g. same color bar across GO_CC/GO_MF/GO_BP). Alternatively, provide a numeric vector of 2 values to manually define lower/upper-limits.

score_color_updown

array of 2 strings that describe the colors for up- and down-regulation (used when score_color is set to "updown"). Default color-coding; up = red, down = blue. Use hex color codes, e.g. "#ff0000" for red.

max_ngenes

only plot terms with less than N genes (quick way to get rid of large/unspecific terms)

topn

topn terms to show after sorting genesets by p-value. For example, this makes it easy to plot the top10 GO terms. Set to NA to ignore this filter (default)

padj_cutoff

adjusted pvalue cutoff for terms shown in the plot. If set to NA (default), all significant genesets are used (i.e. 'signif' column in the input geneset table)

Value

if output_dir is NA, a list of ggplot2 objects. Otherwise, write plots to file and do not return any value

Examples

# note; this example downloads data when first run, and typically takes ~60seconds

# store the downloaded files in the following directory. Here, the temporary file
# directory is used. Alternatively, consider storing this data in a more permanent location.
# e.g. output_dir="~/data/go" on unix systems or output_dir="C:/data/go" on Windows
output_dir = tempdir()

## first run the default example from test_genesets() to obtain geneset results
datasets = download_goat_manuscript_data(output_dir)
genelist = datasets$`Wingo 2020:mass-spec:PMID32424284`
genesets_asis = download_genesets_goatrepo(output_dir)
genesets_filtered = filter_genesets(genesets_asis, genelist)
result = test_genesets(genesets_filtered, genelist, method = "goat",
  score_type = "effectsize", padj_method = "bonferroni", padj_cutoff = 0.05)

# generate lollipop charts for each GO domain (CC/BP/MF), with geneset -log10
# adjusted p-value on the x-axis and color-coding by geneset up/down-regulation
plot_lollipop(
  result, output_dir, plot_type = "lollipop", topn = 50,
  score_xaxis = "minlogp", score_color = "updown"
)

# alternatively, as a barplot
plot_lollipop(
  result, output_dir, plot_type = "barplot", topn = 50,
  score_xaxis = "minlogp", score_color = "updown"
)

# alternatively, color-code genesets by enrichment of significant genes using
# parameter `score_color="oddsratio"` . See further `score_geneset_oddsratio`
# function documentation for definition/computation of this score.
plot_lollipop(
  result, output_dir, plot_type = "lollipop", topn = 50,
  score_xaxis = "minlogp", score_color = "oddsratio"
)