biogridpy

biogridpy is a Python client for the BioGRID webservice.

Installation

To install, clone the repository

$ git clone https://github.com/arvkevi/biogridpy.git
$ cd biogridpy
$ python setup.py

Usage

Before using the client, get a free access key here or if you already have one, see below:

Copy and paste the access key to the existing biogridpyrc file, or you can create a new biogridpyrc file anywhere you like.

$ vi biogridpyrc

biogridpyrc should look like this:

[BioGRID_ak]
access_key = YourAccessKeyHere

Instantiate the BioGRID client:

>>> from biogridpy.biogrid_client import BioGRID
>>> BG = BioGRID(config_filepath='/path/to/biogridpyrc') # provide path to biogridpyrc file

The interactions endpoint is the most commonly utilized. The first and only positional argument is a string that describes the type of result format. Valid keyword arguments are the same parameters from the parameter list. Any parameter that accepts a list of | separated identifiers can be a Python list type object or a file with one identifier per line.

>>> results = BG.interactions('tab2', geneList=['RB1', 'E2F1'],
                                      evidenceList='examples/evidenceList.list',
                                      includeEvidence='true',
                                      taxId=9606)

There are several attributes to describe and work with the result object:

>>> results.count
	11
>>> results.output_format
	'tab2'
>>> results.endpoint
	'interactions'
>>> results.result #formatted result for downstream use
>>> results.raw_result #unformatted result

Available methods are meant to facilitate downstream analyses or for pipeline development. The export method will save the result object as a file in the proper format according to the format specified previously. The toDataFrame method simply formats the data for easy transformation to a pandas DataFrame object.

>>> results.export(outdir='../examples/example_results', 
                  filename='E2F1_RB1_9606')
>>> import pandas as pd
>>> #you could use the following to transform the results regardless of type
	try: #tab2, extendedTab2, tab1
    	df = pd.read_csv(bg_results.toDataFrame(), sep='\t')
	except IOError as e: #json, jsonExtended
    	df = pd.read_json(bg_results.toDataFrame(), orient='index')

The following non-interaction endpoints are also available:

evidence
identifiers
organisms
version

Use the result attribute to view the response.

>>> evid = BG.evidence()
>>> evid.result[:5] #first 5 elements
	['AFFINITY CAPTURE-LUMINESCENCE',
 	'AFFINITY CAPTURE-MS',
 	'AFFINITY CAPTURE-RNA',
 	'AFFINITY CAPTURE-WESTERN',
 	'BIOCHEMICAL ACTIVITY']

walkthrough

The IPython notebook walkthrough.ipynb provides in-depth examples of how to use each method.

$ cd biogridpy
$ jupyter notebook notebooks/walkthrough.ipynb

$ cd biogridpy
$ ipython notebook notebooks/walkthrough.ipynb

Citation

Winter AG, Wildenhain J, Tyers M. BioGRID REST Service, BiogridPlugin2 and BioGRID WebGraph: new tools for access to interaction data at BioGRID. Bioinformatics, 2011 Apr 1.

serval other BioGRID publications