diversity: Shannon diversity index

In MetabolSSMF: Simplex-Structured Matrix Factorisation for Metabolomics Analysis

Shannon diversity index

Description

Calculate the Shannon diversity index of the memberships of an observation. The base of the logarithm is 2.

Usage

diversity(x, two.power = FALSE)

Arguments

x	A membership vector.
two.power	Logical, whether return to the value of 2^{\mathrm{E}(h_{i})}.

Details

Given a membership vector of the i^{th} observation h_i, the Shannon diversity index is defined as

\mathrm{E}(h_{i}) = -\sum_{r=1}^k h_{ir} \mathrm{log}_2 (h_{ir}).

Specifically, in the case of h_{ir}=0, the value of h_{ir} \mathrm{log}_2 (h_{ir}) is taken to be 0.

Value

A numeric value of Shannon diversity index \mathrm{E}(h_{i}) or 2^{\mathrm{E}(h_{i})}.

Author(s)

Wenxuan Liu

Examples

# Memberships vector
membership1 <- c(0.1, 0.2, 0.3, 0.4)
diversity(membership1)
diversity(membership1, two.power = TRUE)

# Memberships matrix
membership2 <- matrix(c(0.1, 0.2, 0.3, 0.4, 0.3, 0.2, 0.4, 0.1, 0.2, 0.3, 0.1, 0.4),
                      nrow=3, ncol=4, byrow=TRUE)

E <- rep(NA, nrow(membership2))
for(i in 1:nrow(membership2)){
  E[i] <- diversity(membership2[i,])
}
E

MetabolSSMF documentation built on April 3, 2025, 5:44 p.m.