Simulate Trees with Separately Controlled Imbalance and Branching Times • treeshapesim

Simulate rooted, binary, ultrametric trees in which topological imbalance and the distribution of branching times are set by separate parameters.

Topology: Aldous’ (1996) beta-splitting model. beta = 0 gives Yule trees, lower values give more imbalanced trees and higher values more balanced ones.
Branching times: a pure-birth process in which each of the k lineages splits at rate k^c. c = 0 is Yule, c < 0 slows diversification towards the present (negative gamma; Pybus & Harvey, 2000) and c > 0 speeds it up.

The gamma statistic depends only on the internode intervals and the Colless index only on the topology, so beta mainly moves the Colless index and c mainly moves gamma.

Documentation: https://diogoprov.github.io/treeshapesim/

Installation

install.packages("treeshapesim", repos = c("https://diogoprov.r-universe.dev",
                                           "https://cloud.r-project.org"))
# or
remotes::install_github("diogoprov/treeshapesim")

Example

library(treeshapesim)
set.seed(1)
tr <- sim_tree_shape(100, beta = -1, c = -0.5)   # imbalanced, early splits
tree_shape(tr)                                    # gamma and normalized Colless

# trees near target values of gamma and Colless
cal <- calibrate_shape(100)                       # calibration for 100 tips
trees <- sim_tree_target(100, gamma = -2, colless = 1, nsim = 50, tol = 0.25,
                         calibration = cal)

# a 3 x 3 grid of tree shapes and the parameters of each cell
grid <- shape_grid(100, gamma = c(-2, -1, 0), colless = c(0, 0.5, 1),
                   calibration = cal)

See vignette("treeshapesim") for a walk-through. Many other tree statistics can be computed with treestats.

How treeshapesim differs from TreeSim and apTreeshape

	treeshapesim	TreeSim	apTreeshape
Purpose	Trees whose imbalance and distribution of branching times vary independently, for simulation studies on tree shape	Trees from birth–death processes	Analysis and simulation of tree topologies
Topology	Aldous’ beta-splitting model, with `beta` as a continuous parameter	Not a parameter: it follows from the birth–death process	Yule, PDA, Aldous’ branching, biased and user-defined split models (`rtreeshape()`); beta-splitting with a clade age–richness index (`simulate_tree()`, `simulate_yule()`)
Branching times	Pure birth with per-lineage rate k^c, drawn independently of the topology, so gamma can be negative, near zero or positive	Birth–death with constant rates, rate shifts, mass extinctions or density-dependent speciation, conditioned on the number of tips and/or the age	None (`rtreeshape()`), ranks 1 to n − 1 (`simulate_tree()`) or a constant-rate birth–death process (`simulate_yule()`)
Separate control of imbalance and branching times	Yes: the realized gamma depends only on `c` and the realized Colless index only on `beta`	No: imbalance is not set	Imbalance is set by `beta`; branching times have no shape parameter of their own
Target values of gamma and Colless	Yes: calibration, rejection to a tolerance and grids of shapes (`sim_tree_target()`, `shape_grid()`)	No	No
Extinction, serial sampling, gene trees	No (pure birth, complete sampling)	Yes	Extinction in the ranked-tree models
Estimation and tests of tree shape	No	No	Yes (e.g. `maxlik.betasplit()`, `aldous.test()`)

Use TreeSim for trees from specific birth–death models and apTreeshape to model, estimate or test topological imbalance. Use treeshapesim when the imbalance and the distribution of branching times have to be varied as separate factors. The comparison is based on the documentation of TreeSim 2.4 and apTreeshape 1.5-0.1; apTreeshape was archived on CRAN in 2023 and is available from the CRAN archive.

Citation

Provete, D. B. (2026). treeshapesim: Simulate trees with separately controlled imbalance and branching times. R package. https://doi.org/10.5281/zenodo.23647260

This DOI always points to the latest version; each release also has its own DOI on Zenodo. In R, citation("treeshapesim") gives the entry with the installed version.

References

Aldous, D. (1996). Probability distributions on cladograms. In D. Aldous & R. Pemantle (Eds.), Random discrete structures (pp. 1–18). Springer. https://doi.org/10.1007/978-1-4612-0719-1_1

Pybus, O. G., & Harvey, P. H. (2000). Testing macro-evolutionary models using incomplete molecular phylogenies. Proceedings of the Royal Society B, 267, 2267–2272. https://doi.org/10.1098/rspb.2000.1278