Skip to content

A collection of useful python functions for looking up information and working with the Biolink Model

License

Notifications You must be signed in to change notification settings

di2ag/biolink-model-toolkit

 
 

Repository files navigation

Python PyPI

Biolink Model Toolkit

Biolink Model Toolkit (BMT): A Python API for working with the Biolink Model.

BMT provides utility functions to look up Biolink Model for classes, relations, and properties based on Biolink CURIEs or external CURIEs that have been mapped to Biolink Model.

Note: Each release of BMT is pinned to a specific version of the Biolink Model to ensure consistency.

Installation

For users

BMT is available on PyPI and can be installed via pip:

pip install bmt

For developers

You can install BMT from GitHub as follows:

# clone the repo
git clone https://github.com/biolink/biolink-model-toolkit

# cd into the folder
cd biolink-model-toolkit

# install
python setup.py install

Documentation

The technical documentation for BMT can be found at https://biolink.github.io/biolink-model-toolkit/

Usage

BMT provides convenience methods to operate on the Biolink Model.

Using this toolkit you can,

  • Get Biolink Model elements corresponding to a given Biolink class or slot name
  • Get Biolink Model elements corresponding to a given external CURIE/IRI
  • Get ancestors for a given Biolink Model element
  • Get descendants for a given Biolink Model element
  • Get parent for a given Biolink Model element
  • Get children for a given Biolink Model element
  • Check whether a given Biolink Model element is part of a specified subset

Using the Toolkit class

The main entrypoint is the Toolkit class that provides various methods for accessing and working with the Biolink Model.

Getting a Biolink Model element based on its name

from bmt import Toolkit
t = Toolkit()
element = t.get_element('gene') # This returns the element for 'biolink:Gene'

Get a Biolink Model element based on its mappings

from bmt import Toolkit
t = Toolkit()
element_name = t.get_element_by_mapping('SEMMEDDB:CAUSES') # This returns 'causes'
element = t.get_element(element_name)

element_name = t.get_element_by_mapping('RO:0002410') # This returns 'causes'
element = t.get_element(element_name)

Get ancestors for a given Biolink Model element

from bmt import Toolkit
t = Toolkit()
ancestors = t.get_ancestors('gene')

The above returns a list of ancestors: ['gene', 'gene or gene product', 'macromolecular machine', 'genomic entity', 'molecular entity', 'biological entity', 'named thing']

Get descendants for a given Biolink Model element

from bmt import Toolkit
t = Toolkit()
descendants = t.get_descendants('gene or gene product')

The above returns a list of descendants: ['gene', 'gene product', 'gene product isoform', 'RNA product', 'noncoding RNA product', 'microRNA', 'RNA product isoform', 'transcript', 'protein', 'protein isoform']

Check whether a given string is a valid Biolink Model category

from bmt import Toolkit
t = Toolkit()
t.is_category('gene') # True
t.is_category('treats') # False

Check whether a given string is a valid Biolink Model predicate

from bmt import Toolkit
t = Toolkit()
t.is_predicate('related to') # True
t.is_predicate('interacts with') # True
t.is_predicate('disease') # False

Using the Toolkit class with different versions of Biolink Model

BMT is pinned to a specific version of Biolink Model at each release. This can be configured to use your custom version of Biolink Model YAML:

from bmt import Toolkit
t = Toolkit('/path/to/biolink-model.yaml')

The path can be a file path or a URL.

About

A collection of useful python functions for looking up information and working with the Biolink Model

Resources

License

Stars

Watchers

Forks

Packages

No packages published

Languages

  • Python 98.8%
  • Other 1.2%