Aggregation of P-value Ranks using a Beta Distribution and Alpha Cutoff

Description

This function is called internally as a single instance of the beta aggregation step in RRAa. Users should not interact with it directly. The expected input is a list of rank statistics, and a paired alpha argument defining which values to consider in downstream analyses (see below).

Usage

1
ct.RRAalpha(p, g.key, alpha, shuffle = FALSE, return.obj = TRUE)

Arguments

p

A single column matrix of rank statistics, with row.names indicating the gRNA labels.

g.key

data.frame with guide and gene names

alpha

The alpha cutoff parameter, corresponding to the P-value threshold or fold change proportion at which gRNAs should no longer be considered to be differentially expressed. Alternatively, this can be provided as a logical vector of the same length as the number of rows in p, containing only TRUE and FALSE elements indicating whether the element should be included during the aggeregation step.

shuffle

Logical indicating whether to shuffle the rank statistics prior to calculating the rho statistics (useful for permutation).

return.obj

Name of the environment to record results in, or TRUE to return the RRAa results directly as a numeric vector of genewise rho statistics. If an environment is supplied, the function will directly increment the target.positive.iterations variable within it on the basis of the obs variable in the specified environment.

Value

Nothing, or a named list of target-level P-values, which are treated as a rho statistic in the permutation step.

Author(s)

Russell Bainer

Examples

1
2
3
data('fit')
data('ann')
geneScores <- ct.RRAalpha(fit$p.value, ann, alpha = 0.1, shuffle = FALSE, return.obj = TRUE)

Want to suggest features or report bugs for rdrr.io? Use the GitHub issue tracker.