Getting started¶
Requirements¶
- Python 3.10 or later
- A rooted tree in Newick format
Install¶
pip install phytclust
From source:
git clone https://github.com/schwarzlab-ccb/PhytClust.git
cd PhytClust
pip install -e ".[dev]"
Confirm the install with phytclust --version.
First run¶
Cluster the bundled sample tree into five groups:
phytclust examples/sample_tree.nwk --k 5 --save-fig --out-dir results/quickstart
results/quickstart/ then contains:
phytclust_results.tsv— which leaf is in which clustertree_k5.png— the tree coloured by cluster, one colour per cladescores.png— the score-versus-k curve
When k is unknown¶
--top-n 3 returns the best few k values; --resolution returns one k per
scale. The CLI tutorial covers both.
Rooting¶
PhytClust requires a rooted tree. An unrooted tree can be rooted at run time:
phytclust tree.nwk --k 5 --root-taxon "species_A" # on a named taxon
phytclust tree.nwk --k 5 --root-taxon midpoint # midpoint rooting
To drop a taxon before clustering rather than root on it, use --outgroup.
Web GUI (experimental)¶
pip install "phytclust[gui]"
phytclust gui
This opens http://127.0.0.1:8000, where a Newick string can be pasted and explored interactively. Use the CLI or the Python API for reproducible analysis.