plot_combn_response: Plots drug response for scored paired data.

In HenryWard/orthrus: Scoring Combinatorial CRISPR Screens

Plots drug response for scored paired data.

Description

Pretty-plots response for data which uses a derived null model (e.g. for comparing dual-gene knockout effects to a multiplicative null model derived from single-gene effects). Assumes that data was scored by score_combn_vs_single and significant effects were called by call_combn_hits. Writes both a scatterplot and a volcano plot to file in the specified folder.

Usage

plot_combn_response(
  scores,
  condition_name,
  output_folder,
  filter_names = NULL,
  loess = TRUE,
  neg_type = "Negative",
  pos_type = "Positive",
  volcano_type = "FDR",
  plot_type = "png"
)

Arguments

scores	Dataframe of scores returned from call_combn_hits.
condition_name	Name of condition passed to call_combn_hits.
output_folder	Folder to output plots to.
filter_names	If a list of column names is given, calls points as non-significant if they are significant in the provided columsn (e.g. to remove points significant in control screens; default NULL).
loess	If true and data was loess-normalized, plots loess null model instead (default TRUE).
neg_type	Label for significant effects with a negative differential effect passed to call_combn_hits (default "Negative").
pos_type	Label for significant effects with a positive differential effect passed to call_combn_hits (default "Positive").
volcano_type	One of "pval" or "FDR" to specify whether to plot -log10(pval) or -log2(FDR), respectively (default "FDR").
plot_type	Type of plot to output, one of "png" or "pdf" (default "png").

HenryWard/orthrus documentation built on June 2, 2023, 10:28 p.m.