Integral Analysis of Diversity Based on Hill Numbers

Measures and compares the diversity of biological communities (e.g. tables of operational taxonomic units (OTUs), amplicon sequence variants (ASVs) or metagenome-assembled genomes (MAGs)) based on Hill numbers, in a unified framework for neutral, phylogenetic and functional diversity measurement, diversity partitioning, (dis)similarity measurement, diversity profiles, evenness and redundancy. The statistical framework encompasses richness, Shannon and Simpson diversity, Faith's phylogenetic diversity (PD), Rao's quadratic entropy and Sorensen- and UniFrac-type dissimilarities, all grounded in a single Hill-number framework. Methods are described in Jost (2007) , Chao et al. (2010) , Chiu et al. (2014) and reviewed in Alberdi & Gilbert (2019) . Optional import adapters interoperate with the Bioconductor packages 'phyloseq', 'SummarizedExperiment' and 'TreeSummarizedExperiment', which are available from < https://bioconductor.org>.

hilldiv3

hilldiv3 measures and compares the diversity of biological communities — OTU, ASV or MAG count tables — using Hill numbers. Hill numbers are a single, intuitive family of diversity metrics: each one is an effective number of taxa ("how many equally-abundant taxa would give this much diversity"), and one parameter, the diversity order q, slides smoothly between counting all taxa equally (richness), weighting them by abundance (Shannon) and focusing on the common ones (Simpson). Because everything is expressed in the same currency, results are directly comparable across samples, studies and methods.

From that one foundation, hilldiv3 provides a unified toolkit for neutral diversity (abundances only), phylogenetic diversity (accounting for how related taxa are) and functional diversity (accounting for how different their traits are), covering measurement, partitioning, (dis)similarity, profiles, evenness and redundancy. You call the same functions for all three — the diversity type is chosen by whether you supply a tree or a distance matrix.

Installation

# install.packages("devtools")
devtools::install_github("alberdilab/hilldiv3")

Quick start

library(hilldiv3)

# Bundled simulated gut-microbiome MAG data.
hilldiv(gut_counts)                    # neutral Hill numbers q = 0, 1, 2
hilldiv(gut_counts, tree = gut_tree)   # neutral + phylogenetic

dist <- traits2dist(gut_traits)
hilldiv(gut_counts, dist = dist)                   # neutral + functional
hilldiv(gut_counts, tree = gut_tree, dist = dist)  # all three types at once

# Results are tidy by default and plot directly.
plot(hillprof(gut_counts))             # diversity profile
hilldiv(gut_counts, out = "matrix")    # matrix: samples x q orders

Documentation

Full documentation lives on the package website: https://alberdilab.github.io/hilldiv3/

Get started — vignette("hilldiv3"), a gentle introduction for anyone using Hill numbers for the first time.
Articles — step-by-step guides to diversity types, partitioning & (dis)similarity, profiles/evenness/redundancy, and preparing your data.
Examples — complete worked analyses (bat diets, gut microbiomes).
Reference — every exported function, grouped by task.

What's new in v3

If you have used hilldiv2, v3 is a full redesign that keeps the familiar function names (hilldiv(), hillpart(), hilldiss(), hillsim(), hillpair(), hillred(), tss(), traits2dist()) while changing how they work underneath:

A tested, isolated compute engine — the diversity maths lives in one place and is unit-tested independently of the user-facing functions.
A single validation/alignment layer that reorders data to match the tree or distance matrix (fixing silent misalignment in v2), plus a real match_data() helper.
Broad input support: matrices, data frames, tibbles, phyloseq and TreeSummarizedExperiment objects.
Faster phylogenetic computation using an ape post-order traversal in place of geiger::tips(); hillpair() computes the shared structure once and reuses it across all sample pairs.
Tidy by default: every hill* function returns a long-format data.frame with print()/plot()/autoplot() methods; pass out = "matrix" for a plain matrix.
An explicit type = c("auto", "neutral", "phylogenetic", "functional") argument that asserts and validates the diversity type (auto-detected by default).
New functions hillprof() (diversity profiles) and hilleven() (evenness), hierarchical multi-scale partitioning in hillpart(), plus bundled example data (gut_counts, gut_tree, gut_traits).

See NEWS.md for the full changelog.

References

Hill, M.O. (1973). Diversity and evenness. Ecology, 54, 427-432.
Jost, L. (2007). Partitioning diversity into independent alpha and beta components. Ecology, 88, 2427-2439.
Chao, A., Chiu, C.-H. & Jost, L. (2010). Phylogenetic diversity measures based on Hill numbers. Phil. Trans. R. Soc. B, 365, 3599-3609.
Chiu, C.-H., Jost, L. & Chao, A. (2014). Phylogenetic beta diversity, similarity, and differentiation measures based on Hill numbers. Ecological Monographs, 84, 21-44.
Alberdi, A. & Gilbert, M.T.P. (2019). A guide to the application of Hill numbers to DNA-based diversity analyses. Mol. Ecol. Resour., 19, 804-817.