Skip to content

gpvolve.simulate

Population-dynamics simulation backends.

  • wright_fisher(graph, *, fitness_column, population_size, mutation_rate, n_generations, initial_index=0, seed=None) -> NDArray[int64] Discrete-generation multinomial sampling. Returns (n_generations + 1, n_states) of per-generation genotype counts.

  • gillespie_walk(graph, *, fitness_column, source, targets, fixation="sswm", mu_per_step=1.0, max_steps=10_000, seed=None, **fixation_params) -> tuple[NDArray[float64], NDArray[int64]] Exact Gillespie waiting-time trajectory under the chosen fixation kernel. Returns (times, states).

  • slim_available() -> bool

  • tskit_available() -> bool

The SLiM and tskit backends are stubs in v2.0.0; install with pip install 'gpvolve-v2[sim]' to make the availability probes return True. The full SLiM integration is on the roadmap.