Calculate Tsallis entropy for a vector of transcript-level expression values of one gene.

Usage

calculate_tsallis_entropy(
  x,
  q = 2,
  norm = TRUE,
  what = c("S", "D", "both"),
  log_base = exp(1)
)

Arguments

x: Vector of (non-negative) expression values.
q: Tsallis entropy parameter (q > 0). Scalar or numeric vector (default: 2).
norm: Logical; if TRUE, normalize entropy by its theoretical maximum (values in [0,1]).
what: Which quantity to return: 'S' (Tsallis entropy), 'D' (Hill numbers), or 'both'.
log_base: Base of the logarithm used for Shannon limits and normalization (default: exp(1)).

Value

For `what = 'S'` or `what = 'D'`: a numeric vector (named when length(q) > 1). For `what = 'both'`: a list with components `$S` and `$D`.

Details

Implements S_q = (1 - sum p^q)/(q - 1) and D_q = (sum p^q)^(1/(1-q)). Uses the q->1 limits (Shannon entropy and its exponential). Natural logarithms are used for the q->1 limit and for normalization.

Examples

x <- c(10, 5, 0)
calculate_tsallis_entropy(x, q = c(0.5, 1, 2), norm = TRUE)
#>     q=0.5       q=1       q=2 
#> 0.5380048 0.5793802 0.6666667