What problem does it solve?
Phylogenetic analysis requires computing dozens of tree and alignment statistics (treeness, RCV, DVMC, parsimony-informative sites, saturation, long-branch scores) across hundreds of ortholog files, then running group comparisons like Mann-Whitney U tests. Doing this manually with PhyKIT per file is slow, error-prone, and easy to get wrong by parsing the wrong output column or processing only a sample of genes.
Core Features & Use Cases
- Batch PhyKIT metrics: Compute treeness, saturation, DVMC, evolutionary rate, tree length, long-branch score, patristic distances, RCV, gap percentage, and parsimony-informative sites across entire directories of alignments and trees.
- BUSCO scogs pipelines: Bundled scripts extract pre-computed alignments and trees from
scogs_animals.zip / scogs_fungi.zip, apply single-copy ortholog intersection rules, and emit per-group summaries plus Mann-Whitney U statistics in one pass.
- De novo alignment and tree building: Align raw sequences with Clustal Omega, MUSCLE, or MAFFT via EBI_msa_align, then build neighbour-joining or UPGMA trees with EBI_build_phylogenetic_tree.
- Use Case: Given a data folder of fungal and animal ortholog trees, answer "what is the median DVMC for fungi and the Mann-Whitney U statistic comparing fungi vs animals" with a single script invocation.
Quick Start
Ask the agent to compute the median treeness across all tree files in your data folder and compare the fungi and animals groups with a Mann-Whitney U test.