Manhattan plot with all tested snarls

Generate a Manhattan plot showing the p-values across the pangenome.

Usage

manhattan_plot(
  assoc,
  p_threshold = 1e-05,
  output_file = NULL,
  chr_order = NULL,
  show_all_points = FALSE,
  show_top_points = 0,
  wrap_by_rows = 1,
  no_plotting = FALSE
)

Arguments

assoc

Either the data.frame imported by *import_assoc*, or the path to STOAT's output (*assoc.pvalues.tsv.gz)

p_threshold

P-value threshold for the horizontal significance line (default: 1e-5).

output_file

If not NULL, the name of the output image where to save the plot (image type guessed from the file name).

chr_order

list of chromosome names in the desired order for the plot. If NULL, will try to guess.

show_all_points

should we plot all the points? Default is FALSE (cluster close-by points instead). TRUE is slower (and heavier in PDF) for large genomes.

show_top_points

how many top associations should be shown as single points? Default is 0.

wrap_by_rows

how many rows to wrap the chromosome panels. Default: 1 (no wrapping)

no_plotting

if TRUE, no graph is plotted and the ggplot/gtable is just returned. Only useful in very specific situations, with multi-row plots and where you want to handle the plotting yourself.

Value

A ggplot object. Saves a file if output_file is provided.

Examples

# prepare the filename
assoc_file = system.file('extdata/stoat.quantitative.assoc.pvalues.tsv.gz', package='StoatPlot')

# manhattan plot from the filename (will load the file internally)
manhattan_plot(assoc_file, show_top_points=1000)


# -------------------------------------

# import the association results first
assoc = import_assoc(assoc_file)

# then do a Manhattan plot (recommended for large dataset).
# Also change the P-value threshold
manhattan_plot(assoc_file, p_threshold=1e-3)