ct.CAT: Compare Two CRISPR Screens via a CAT plot

In gCrisprTools: Suite of Functions for Pooled Crispr Screen QC and Analysis

Description

This is a function for comparing the results of two screening experiments. Given two summaryDF, the function places them in register with one another, generates a Concordance At The Top (CAT) plot, and returns an invisible dataframe containing the relevant gene-level signals.

This function is conceptually similar to the 'ct.ROC' and 'ct.PRC()' functions, but is appropriate when considering consistency of ranked values rather than an interchangeable set; the most common use case is for comparing primary and replication screens, where the underlying technology and selection criteria are expected to be highly similar. CAT plots are fundamentally about comparing rankings, and so only targets in common between the two provided screens are considered. If the totality of list overlap is important, consider using 'ct.PRC()' or 'ct.ROC()'.

Note that ranking statistics in CRISPR screens are (usually) permutation-based, and so some granularity in the rankings is expected. This function does a little extra work to limit the influence of this granularity and to ensure that hits are counted as soon as the requisite value of the ranking statistic is reached regardless of where the target is located within the block of equally-significant hits. Functionally, this means that the drawn curve is somewhat anticonservative in cases where the target ranks are not well differentiated.

Usage

ct.CAT(
  df1,
  df2,
  targets = c("geneSymbol", "geneID"),
  enrich,
  plot.it = TRUE,
  plot.rho = TRUE
)

Arguments

df1	A dataframe summarizing the results of the screen, returned by the function ct.generateResults.
df2	A dataframe summarizing the results of the screen, returned by the function ct.generateResults.
targets	Column of the provided summaryDF to consider. Must be geneID or geneSymbol.
enrich	Logical indicating whether to test for enrichment or depletion.
plot.it	Logical indicating whether to compose the plot on the default device.
plot.rho	Logical indicating whether to plot the Rho values in addition to the P-values, which sometimes have better ranking properties.

Value

Invisibly, a data.frame containing the relevant summary stats for each target in both screens.

Author(s)

Russell Bainer

Examples

data('resultsDF')
cat <- ct.CAT(resultsDF, resultsDF[1:2000,], enrich = TRUE)
head(cat)

gCrisprTools documentation built on Nov. 8, 2020, 8:17 p.m.