Calculate Experimental P-Value Based on Bootstrap Contributions

For a single signature, calculate the proportion of bootstraps that have a percentage contribution below a specified threshold. This value represents an 'experimental' P-value. For example, if P < 0.05, there is a <5% chance of the signature having a contribution >= threshold by chance.

Usage

sig_compute_experimental_p_value(bootstraps, threshold)

Arguments

bootstraps

A numeric vector describing the percentage contributions of a single signature in different bootstraps.

threshold

A numeric value specifying the minimum contribution threshold (typically 0.05). This is referred to as the sparsity threshold.

Value

The proportion of bootstraps with contributions below threshold. This represents an experimental p-value (referred to as the sparsity p-value).

Examples

bootstraps <- c(0.01, 0.03, 0.07, 0.02, 0.05)
threshold <- 0.05
sig_compute_experimental_p_value(bootstraps, threshold)
#> [1] 0.6