phypanda 2.1.0

a package for computing and maximizing phylogenetic diversity in phylogenetic networks

phypanda

phypanda is the Python library for PaNDA—an algorithm library for phylogenetic diversity (PD) optimization on directed phylogenetic networks. It builds on phylozoo for network representations and scanwidth for node- and edge-scanwidth–based dynamic programming.

Key Features

• All-paths (MAPPD), Max-tree and Min-tree PD: compute the diversity score of a fixed taxon set for several PD measures, or maximize PD using exact FPT algorithms parameterized by node scanwidth or edge scanwidth.
• Budgeted maximization: assign integer costs to taxa and maximize PD under budget constraints.
• JIT-compilation: speed up algorithms by optional numba JIT compilation.
• High-level API — compute_diversity, marginal_diversities, greedy_max_diversity, and solve_max_diversity work with any measure and dispatch to the appropriate solver.

Installation

pip install phypanda

Runtime dependencies (phylozoo, networkx, scanwidth, numba) are installed automatically. For development or documentation extras:

pip install phypanda[dev]   # testing
pip install phypanda[docs]  # Sphinx documentation

Documentation

For the full manual, API reference, and installation guide, visit the phypanda docs.

Citation

If you use phypanda, please cite:

Niels Holtgrefe, Leo van Iersel, Ruben Meuwese, Yukihiro Murakami, and Jannik Schestag. PaNDA: Efficient Optimization of Phylogenetic Diversity in Networks. bioRxiv, 2025. doi: 10.1101/2025.11.14.688467

If your work builds specifically on the budgeted node-scanwidth algorithms, please also cite:

Niels Holtgrefe and Jannik Schestag. Tractable Optimization of Budgeted Phylogenetic Diversity on Networks Utilizing Node-Scanwidth. 2026.

See also

For the graphical-user interface developed for the first paper, please go to gui/.

For the experimental materials corresponding to the above two papers, please go to experiments/.