harryhaller001 / keggtools

Library for KEGG pathway enrichment analysis

keggtools

Library for KEGG pathway enrichment analysis.

Documentation: https://keggtools.org/

PyPi: https://pypi.org/project/keggtools/

Installation

keggtools only supports python version greater than or equal 3.8.

Dependencies

• graphviz

python dependencies

• requests
• pydot
• scipy

Installation keggtools package using pip:

python3 -m pip install keggtools

To get a more detailed list of install options, please read the INSTALL.md

API

Download and Parsing

from keggtools import Resolver, IMMUNE_SYSTEM_PATHWAYS

ORGANISM_ID = "hsa"
resolver = Resolver()

# Select first immune system pathway as example
pathway_id = list(IMMUNE_SYSTEM_PATHWAYS.keys())[1]

# Resolve pathway
pathway = resolver.get_pathway(organism=ORGANISM_ID, code=pathway_id)
print(pathway)

Enrichment and Testing

from keggtools import Enrichment

# Add pathway object to list
pathway_list = []

# Init analysis with organism code
analysis = Enrichment(pathways=pathway_list)

# Study genes as list of entrez gene id's
study_genes = []
analysis.run_analysis(gene_list=study_genes)

# to_dataframe method requires pandas installation
result = analysis.to_dataframe()
print(result.head())

Rendering

from keggtools.render import Renderer

# Load and parse pathway
renderer = Renderer(kegg_pathway=pathway)

# Render to dot graph
renderer.render()

# Export to png
renderer.to_file("output.png", extension="png")

Development

Dev installation

Fast install with virtualenv for development.

python3 -m virtualenv venv
source ./venv/bin/activate
pip install --upgrade pip

# Install from requirements
pip install -r requirements.txt

# Or use makefile
make install

Linting

Run linting with pylint for keggtools package.

# Run linting for package
make pylint

Static code analysis

Static code analysis using mypy. Run static code analysis with mypy.

# Run static code analysis
make mypy

Testing

Run unittest for keggtools package.

# Run unittest for package
make pytest

Install package from repo

The package is using the flit backend with a pyproject.toml and twine. To install from repo use

# Install package from repo
make twine