orthoxml-tools

Tools for working with OrthoXML files.

Table of contents

• What is OrthoXML?
• Installation
• Quick start
• Common workflows
  • Inspect an OrthoXML file
  • Export data for downstream analysis
  • Transform or subset data
  • Convert between formats
• Command reference
• Getting help
• Legacy API
• Development and testing
• Citation

What are OrthoXML and OrthoXML-tools?

OrthoXML is a standard for sharing and exchanging orthology predictions. It provides a structured way to describe orthology relationships while preserving database-specific annotations. More details are available at OrthoXML. OrthoXML-tools provide a comprehensive toolkit for working (inspecting, manipulating, convert to/from) with OrthoXML files.

Installation

Install the package from PyPI:

pip install orthoxml-tools

Input OrthoXML files can be plain text or compressed with gzip (.gz) or bzip2 (.bz2).

Quick start

The CLI follows this pattern:

orthoxml-tools [options] <subcommand> [options]

A few common first steps:

# Validate a file against the schema declared inside it
orthoxml-tools validate --infile examples/data/ex1.orthoxml

# Summarize the tree structure
orthoxml-tools stats --infile examples/data/ex3-int-taxon.orthoxml

# Print a human-readable taxonomy tree
orthoxml-tools taxonomy --infile examples/data/ex3-int-taxon.orthoxml

Common workflows

Inspect an OrthoXML file

Use these commands to understand the content of an OrthoXML file before processing it further.

# Basic statistics
orthoxml-tools stats --infile path/to/file.orthoxml

# Validate the file against the OrthoXML Standard
orthoxml-tools validate --infile path/to/file.orthoxml

# Gene counts per taxon
orthoxml-tools gene-stats --infile path/to/file.orthoxml --outfile gene_stats.json

# Find the rootHOG (id, taxonomic level, size) containing given genes
orthoxml-tools find-roothog --infile path/to/file.orthoxml --id protId --genes P12345 Q67890

# Show the taxonomy tree in a human-readable format
orthoxml-tools taxonomy --infile path/to/file.orthoxml

# Write the taxonomy tree to NHX format with annotated internal nodes
orthoxml-tools taxonomy --infile path/to/file.orthoxml --outfile species-tree.nhx

Export data for downstream analysis

Export pairwise relationships or ortholog groups into tabular formats that are easy to analyze with other tools.

# Export orthologous pairs
orthoxml-tools export-pairs ortho \
  --infile examples/data/ex1-int-taxon.orthoxml \
  --outfile orthos.tsv

# Export paralogous pairs
orthoxml-tools export-pairs para \
  --infile examples/data/ex1-int-taxon.orthoxml \
  --outfile paras.tsv

# Export ortholog groups as a simple two-column table
orthoxml-tools export-ogs \
  --infile examples/data/sample-for-og.orthoxml \
  --outfile ogs.tsv \
  --id protId

Transform or subset data

Use these commands to focus on a subset of the tree or to remove incomplete groups.

# Keep only selected species
orthoxml-tools subset \
  --infile examples/data/sample-for-subset.orthoxml \
  --outfile mammals.orthoxml \
  --species "Homo sapiens" "Mus musculus"

# Extract one or more HOGs as standalone root groups
orthoxml-tools subset \
  --infile examples/data/sample-for-subset.orthoxml \
  --outfile opistokonta.orthoxml \
  --hog-ids HOG_Opistokonta

# Filter by completeness score
orthoxml-tools filter \
  --infile examples/data/sample-for-filter.orthoxml \
  --threshold 0.24 \
  --strategy cascade-remove \
  --outfile filtered.orthoxml

Convert between formats

The package also supports conversions to and from common phylogenetic and orthology formats.

# Convert OrthoXML to NHX trees
orthoxml-tools to-nhx \
  --infile examples/data/sample-for-nhx.orthoxml \
  --outdir ./tests_output/trees \
  --xref-tag protId \
  --encode-levels

# Convert NHX back to OrthoXML
orthoxml-tools from-nhx \
  --infile examples/data/sample.nhx \
  --outfile ./tests_output/from_nhx.orthoxml

# Convert OrthoFinder-style CSV to OrthoXML
orthoxml-tools from-csv \
  --infile examples/data/InputOrthogroups.csv \
  --outfile ./tests_output/orthofinder.orthoxml

# Convert OrthoXML back to OrthoFinder-style CSV
orthoxml-tools to-csv \
  --infile  ./tests_output/orthofinder.orthoxml \
  --outfile examples/data/InputOrthogroups.csv

Command reference

validate

Validate an OrthoXML file against the schema version declared in the file.

orthoxml-tools validate --infile path/to/file.orthoxml

Options:

• --infile <file>: Input OrthoXML file (required).

stats

Display basic tree statistics.

orthoxml-tools stats --infile path/to/file.orthoxml

gene-stats

Display gene counts per taxon.

orthoxml-tools gene-stats --infile path/to/file.orthoxml [--outfile <file>]

Options:

• --infile <file>: Input OrthoXML file (required).
• --outfile <file>: Write counts to a JSON file when provided.

taxonomy

Print a taxonomy tree for the provided OrthoXML file.

orthoxml-tools taxonomy --infile path/to/file.orthoxml

Options:

• --infile <file>: Input OrthoXML file (required).
• --outfile <file>: When provided, write the taxonomy tree to this file in NHX format with internal nodes annotated.

find-roothog

Find the rootHOG that contains each of the given genes and report its ID, taxonomic level and size.

orthoxml-tools find-roothog --infile path/to/file.orthoxml \
  [--genes GENE [GENE ...]] \
  [--genes-file FILE] \
  [--id <id|protId|geneId>] \
  [--outfile <file>]

Options:

• --infile <file>: Input OrthoXML file (required).
• --genes <id> [<id> ...]: One or more gene identifiers to look up.
• --genes-file <file>: Plain-text file with one gene identifier per line.
• --id <tag>: The <gene> attribute the identifiers refer to: the internal OrthoXML id (default), or protId / geneId if the file has them. The command stops with an error if no gene in the file has the chosen attribute.
• --outfile <file>: Write the result to this file instead of stdout.

For example, given <gene id="1000000001" protId="A0A8M1N6K4"/>:

orthoxml-tools find-roothog --infile file.orthoxml --genes 1000000001
orthoxml-tools find-roothog --infile file.orthoxml --id protId --genes A0A8M1N6K4

Output is a TSV with the columns query, gene_id (internal OrthoXML id), roothog_id, taxon_level, and num_genes (distinct genes in the rootHOG, including those in nested and paralog groups).

Notes:

• taxon_level is the rootHOG's <property name="TaxRange"> value (OrthoXML 0.3/0.4, e.g. FastOMA output). If that is missing, it is the taxonId attribute (OrthoXML 0.5), shown as the taxon name from <taxonomy> when available.
• Missing values are written as NA. Genes that are absent or not in any group produce an all-NA row and a warning.
• Built for large files: rather than parsing the XML, it searches the raw bytes and only parses the rootHOGs that contain a query. It stops as soon as every query has been found. A 4 GB file is scanned in a few seconds.
• It assumes the group tags have no namespace prefix and that no group is self-closing (<orthologGroup/>), which holds for standard OrthoXML output such as FastOMA's.

export-pairs

Export ortholog or paralog pairs as tab-separated output.

orthoxml-tools export-pairs <ortho|para> \
  --infile <file> \
  --outfile <file> \
  [--id <tag>] \
  [--chunk-size <number>] \
  [--buffer-size <bytes>]

Options:

• --infile <file>: Input OrthoXML file (required).
• --outfile <file>: Output file (required).
• --id <tag>: Identifier to use in the output (id, geneId, or protId).
• --chunk-size <number>: Number of pairs to buffer per write (default: 20,000).
• --buffer-size <bytes>: I/O buffer size in bytes (default: 4 MiB).

export-ogs

Export orthologous groups as a simple TSV file.

orthoxml-tools export-ogs --infile path/to/file.orthoxml --outfile path/to/output.tsv [--id <tag>]

subset

Extract a subset of an OrthoXML file by species and/or HOG IDs.

orthoxml-tools subset --infile path/to/file.orthoxml --outfile path/to/output.orthoxml \
  [--species SPECIES [SPECIES ...]] \
  [--species-file FILE] \
  [--hog-ids HOG_ID [HOG_ID ...]] \
  [--hog-ids-file FILE]

Options:

• --infile <file>: Input OrthoXML file (required).
• --outfile <file>: Output OrthoXML file (required).
• --species <name> [<name> ...]: One or more species names to keep.
• --species-file <file>: Plain-text file with one species name per line.
• --hog-ids <id> [<id> ...]: One or more HOG IDs to extract as new root groups.
• --hog-ids-file <file>: Plain-text file with one HOG ID per line.

Notes:

• HOG IDs can refer to any nesting level; the matched subtree is promoted to a root HOG in the output.
• If both a parent and child HOG ID are supplied, only the parent is extracted.

split

Split the tree into multiple trees based on root HOGs.

orthoxml-tools split --infile path/to/file.orthoxml --outdir path/to/output_folder

filter

Filter the tree by completeness score using a chosen strategy.

orthoxml-tools filter \
  --infile path/to/file.orthoxml \
  --threshold <value> \
  --strategy <cascade-remove|extract|reparent> \
  --outfile path/to/output.orthoxml

to-nhx

Convert OrthoXML to Newick/NHX format.

orthoxml-tools to-nhx --infile path/to/file.orthoxml --outdir path/to/output_folder --xref-tag protId

Options:

• --infile <file>: Input OrthoXML file (required).
• --outdir <folder>: Output folder for generated files (required).
• --xref-tag <tag>: Gene attribute to use as the leaf label (default: protId).
• --encode-levels: Include group-level information as NHX comments.

from-nhx

Convert Newick/NHX files back to OrthoXML.

orthoxml-tools from-nhx --infile path/to/file.nhx --outfile path/to/file.orthoxml [--species-encode nhx|underscore]

from-csv

Convert an OrthoFinder-style CSV file to OrthoXML.

orthoxml-tools from-csv --infile path/to/file.csv --outfile path/to/file.orthoxml

Note: Because the CSV format does not preserve full hierarchical structure, the resulting OrthoXML is reported at the root level and should be considered an exploratory conversion.

to-csv

Export an OrthoXML file to the OrthoFinder-style TSV format.

orthoxml-tools to-csv --infile path/to/file.orthoxml --outfile path/to/file.tsv [--id <attribute>]

Getting help

To see help for any command:

orthoxml-tools --help
orthoxml-tools -h
orthoxml-tools stats --help
orthoxml-tools stats -h

Legacy API

The older object-oriented interface is deprecated and will be removed in v1.0.0. The legacy documentation remains in LEGACY-README.md.

Development and testing

uv install ".[test]"
pytest -vv

# CLI smoke test
tests/test_cli.sh

Citation

If you use orthoxml-tools in your research, please cite:

Yazdizadeh Kharrazi A, Altenhoff AM, Romashchenko N, Dessimoz C, Majidian S. "OrthoXML-Tools: A Toolkit for Manipulating OrthoXML Files for Orthology Data." Journal of Molecular Evolution 2025;93(6):800. https://doi.org/10.1007/s00239-025-10271-7

Publication link: https://link.springer.com/article/10.1007/s00239-025-10271-7