BIOSERVICES: access to biological web services programmatically¶

Python_version_available: BioServices is tested for Python 3.7, 3.8, 3.9, 3.10
Contributions: Please join https://github.com/cokelaer/bioservices
Issues: Please use https://github.com/cokelaer/bioservices/issues
How to cite: Cokelaer et al. BioServices: a common Python package to access biological Web Services programmatically Bioinformatics (2013) 29 (24): 3241-3242
Documentation: RTD documentation.

Bioservices is a Python package that provides access to many Bioinformatices Web Services (e.g., UniProt) and a framework to easily implement Web Services wrappers (based on WSDL/SOAP or REST protocols).

The primary goal of BioServices is to use Python as a glue language to provide a programmatic access to several Bioinformatics Web Services. By doing so, elaboration of new applications that combine several of the wrapped Web Services is fostered.

One of the main philosophy of BioServices is to make use of the existing biological databases (not to re-invent new databases) and to alleviates the needs for expertise in Web Services for the developers/users.

BioServices provides access to about 40 Web Services.

Contributors¶

Maintaining BioServices would not have been possible without users and contributors. Each contribution has been an encouragement to pursue this project. Thanks to all:

https://contrib.rocks/image?repo=cokelaer/bioservices

Quick example¶

Here is a small example using the UniProt Web Service to search for the zap70 specy in human organism:

>>> from bioservices import UniProt
>>> u = UniProt(verbose=False)
>>> data = u.search("zap70+and+taxonomy_id:9606", frmt="tsv", limit=3,
...                 columns="id,length,accession, gene_names")
>>> print(data)
Entry name   Length  Entry   Gene names
ZAP70_HUMAN  619     P43403  ZAP70 SRK
B4E0E2_HUMAN 185     B4E0E2
RHOH_HUMAN   191     Q15669  RHOH ARHH TTF

Note

major changes of UniProt API changed all columns names in June 2022. The code above is valid for bioservices versions >1.10. Earlier version used:

>>> data = u.search("zap70+and+taxonomy:9606", frmt="tab", limit=3,
...                 columns="entry name,length,id, genes")

Note that columns names have changed, the frmt was changed from tab to tsv and taxonomy is now taxonomy_id. Names correspondences can be found in:

u._legacy_names

More examples and tutorials are available in the On-line documentation

Current services¶

Here is the list of services available and their testing status.

Service	CI testing
arrayexpress
bigg
biocontainers
biodbnet
biogrid
biomart
biomodels
chebi
chembl
cog
dbfetch
ena
ensembl
eutils
eva
hgnc
intact_complex
kegg
muscle
mygeneinfo
ncbiblast
omicsdi
omnipath
panther
pathwaycommons
pdb
pdbe
pfam
pride
psicquic
pubchem
quickgo
reactome
rhea
seqret
unichem
uniprot
wikipathway

Note

Contributions to implement new wrappers are more than welcome. See BioServices github page to join the development, and the Developer guide on how to implement new wrappers.

Bioservices command¶

In version 1.8.2, we included a bioservices command. For now it has only one subcommand to download a NCBI accession number and possibly it genbank or GFF file (if available):

bioservices download-accession --accession K01711.1 --with-gbk

Changelog¶

Version	Description
1.11.1	Fix regression i uniprot.mapping (https://github.com/cokelaer/bioservices/issues/245)
1.11.0	Fix uniprot limitation of 25 results only ( For developers: all services are now refactorised to use services as an attribute rather than a parent class. Remove ReactomeOld and ReactomeAnalysis (deprecated) move rnaseq_ebi (deprecated) to attic for book_keeping
1.10.4	Fix v1.10.3 adding missing requirements.txt
1.10.3	Update pdb service to use v2 API remove biocarta (website not accesible anymore) Update Chembl (no API changes)
1.10.2	Fix #226 and applied PR from Fix from @GianArauz https://github.com/cokelaer/bioservices/pull/232 about UniProt error Update MANIFEST to fix #232
1.10.1	allow command line to download genbank and GFF update pride module to use new PRIDE API (July 2022) Fixed KEGG bug #225
1.10.0	Update uniprot to use the new API (june 2022)
1.9.0	Update unichem to reflect new API
1.8.4	biomodels. Fix #208 KEGG: fixed #204 #202 and #203
1.8.3	Eutils: remove warning due to unreachable URL. Set REST as attribute rather and inheritance. NEW biocontainers module KEGG: add save_pathway method. Fix parsing of structure/pdb entry remove deprecated function from Reactome
1.8.2	Fix suds package in code and requirements
1.8.1	Integrated a change made in KEGG service (DEFINITON was changed to ORG_CODE) for developers: applied black on all modules switch suds-jurko to new suds community
1.8.0	add main standalone application. moved chemspider and clinvitae to the attic removed picr service, not active anymore
1.4.X	NEW RNAseq from EBI in rnaseq_ebi module Replaced deprecated HGNC with the official web service from genenames.org Fully updated EUtils since WSDL is now down; implementation uses REST now. Removed the apps/taxonomy module now part of http://github.com/biokit.
1.3.X	CACHE files are now stored in a general directory in the home New REST class to use requests package instead of urllib2. Creation of a global configuration file in .config/bioservice/bioservices.cfg NEW services: Reactome, Readseq, Ensembl, EUtils
1.2.X	NEW services: BioDBnet, BioDBNet, MUSCLE, PathwayCommons, GeneProf
1.1.X	NEW services: biocarta, pfam, ChEBI, UniChem
1.0.0:	first stable release
0.9.X:	NEW services: BioModels, Kegg, Reactome, Chembl, PICR, QuickGO, Rhea, UniProt,WSDbfetch, NCBIblast, PSICQUIC, Wikipath

Warning

Some of the services may be down. BioServices developers are not responsible for the maintenance or failure of underlying services. Generally speaking (and by experience) the services are up most of the time but failure may occur because the site is under maintenance or too many requests have been sent. Another common reason is the fact that the API of the web services has changed: If so, BioServices need to be updated. You may contribute or report such API changes on our Issue page

Installation¶

BioServices is available on PyPi, the Python package repository. The following command should install BioServices and its dependencies automatically provided you have pip on your system:

pip install bioservices

If not, please see the external pip installation page or pip installation entry. You may also find information in the troubleshootings page section about known issues.

Regarding the dependencies, BioServices depends on the following packages: BeautifulSoup4 (for parsing XML), SOAPpy and suds (to access to SOAP/WSDL services; suds is used by ChEBI only for which SOAPpy fails to correctly fetch the service) and easydev. All those packages should be installed automatically when using pip installer. Since version 1.6.0, we also make use of pandas and matplotlib to offer some extra functionalities.