OGER: OntoGene's Biomedical Entity Recogniser

A flexible, dictionary-based system for extracting biomedical terms from scientific articles.

Demo

A demo version of OGER is hosted at https://pub.cl.uzh.ch/purl/OGER.

Installation

Install OGER from its repository using pip:

pip install git+https://github.com/OntoGene/OGER.git

Make sure you use Python 3's pip (eg. pip3). Python 2.x is not supported.
Note: By default, pip installs Python packages at the system level, which typically requires root/admin privileges. To install OGER to a user-owned location, set the --user flag.

Usage: Command-line Tool

The installation process should provide you with an executable oger, which is the common starting point for a number of command-line tools. Type oger CMD, followed by command-specific options, to run the desired tool:

oger run       # run an annotation job
oger serve     # start a REST API server
oger eval      # determine annotation accurateness
oger test      # run software tests
oger version   # print the version number

To see a list of available options for each command, run eg. oger serve -h.

As an alternative to the oger executable, you may run python3 -m oger CMD.

Usage: Python Library

Get config and start a pipeline server:

>>> from oger.ctrl.router import Router, PipelineServer
>>> conf = Router(termlist_path='testfiles/test_terms.tsv')
>>> pl = PipelineServer(conf)

Note: test files can be downloaded here.

Load a text document from disk (using the included test suite):

>>> doc = pl.load_one('testfiles/txt/13373697.txt', 'txt')
>>> doc
<Article with 1 subelement at 0x7f8b2135d860>
>>> doc.text
'[The kind and the measure of ventilation disorders in tuberculous bronchostenosis in relation to its localization]. \n'

Download a collection of articles from PubMed:

>>> coll = pl.load_one(['21436587', '21436588'], fmt='pubmed')
>>> coll
<Collection with 2 subelements at 0x7f8b215f4cc0>
>>> coll[0]
<Article with 2 subelements at 0x7f8b2156a5f8>
>>> coll[0][0]
<Section with 1 subelement at 0x7f8b2156a358>
>>> coll[0][0].text
'Human prostate cancer metastases target the hematopoietic stem cell niche to establish footholds in mouse bone marrow.\n'

Run entity recognition:

>>> pl.process(coll)
>>> entity = next(coll[0].iter_entities())
>>> entity.text, entity.start, entity.end
('Human', 0, 5)
>>> entity.cid
'9606'
>>> entity.info
('organism', 'Homo sapiens', 'NCBI Taxonomy', '9606', 'DC')

Export to disk:

>>> with open('output/collection.json', 'w', encoding='utf8') as f:
...     pl.write(coll, 'bioc_json', f)

The second argument specifies the output format. OGER supports BioC (XML and JSON), PubTator, PubAnnotation JSON, BioNLP stand-off, and CoNLL, among others. A full list of available formats is given here (see the export-format parameter).

Documentation

Documentation is maintained in the GitHub wiki.

Prerequisites

OGER runs on Python 3.4+.

The following third-party libraries need to be installed (pip should take care of this):

• bottle (needed only for the RESTful server)
• lxml
• NLTK

Publications

If you use OGER in an academic context, please cite us:

Lenz Furrer, Anna Jancso, Nicola Colic, and Fabio Rinaldi (2019): OGER++: hybrid multi-type entity recognition. In: Journal of Cheminformatics 11:7. DOI: 10.1186/s13321-018-0326-3 | PDF | bibtex |

Lenz Furrer and Fabio Rinaldi (2017): OGER: OntoGene's Entity Recogniser in the BeCalm TIPS Task. In: Proceedings of the BioCreative V.5 Challenge Evaluation Workshop, pp. 175–182. | PDF | bibtex |

Marco Basaldella, Lenz Furrer, Carlo Tasso, and Fabio Rinaldi (2017): Entity recognition in the biomedical domain using a hybrid approach. In: Journal of Biomedical Semantics 8:51. DOI: 10.1186/s13326-017-0157-6 | PDF | bibtex |

License

OGER offers a dual licensing model.

You can redistribute OGER and/or modify it under the terms of the GNU Affero General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.

The GNU Affero General Public License is designed to ensure that if a modified version is distributed or made accessible on a server (e.g. in a SaaS offering), the modified source code becomes available to the community.

This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Affero General Public License for more details.

You should have received a copy of the GNU Affero General Public License along with this program. If not, see http://www.gnu.org/licenses/.

If you wish to use OGER under alternate terms, you may obtain a commercial license to OGER. Please contact us for more information (http://www.ontogene.org).