seq-cluster-eval

A pipeline for evaluating performance of unsupervised clustering-based species delimitation on simulated eDNA metabarcoding datasets.

Quick Start

1. Download from source:

git clone https://github.com/ComputationalAgronomy/seq-cluster-eval.git

cd seq-cluster-eval

2. Install dependencies:

pip install -r requirements.txt

3. Install package locally:

pip install -e .

4. Verify installation:

pytest # or pytest tests/test_XXX.py

Basic usage

Step	Module	Output	Options
1	TreeSimulator	.nwk file
2	SeqSimulator	.fasta file
3	SeqEncoder	matrix	distance or feature
4	External Pacakges	embedding	PCA / MDS / UMAP
5	External Pacakges	labels	k-means / HDBSCAN
6	Metrics	results DataFrame

See example.

Dependencies

Python packages: See requirements.txt

External software - IQ-TREE: http://www.iqtree.org/doc/AliSim