bioNerDS

Introduction

Welcome to bioNerDS! The bioinformatics named entity recogniser for databases and software.

This tool makes heavy use of GATE (version 6.1). It can be run in sandbox mode, which means a installation of GATE is not a prerequisite, but you will instead need to point the config to a unzipped gate_plugins directory instead (located in the bin/Files directory).

Results from full-text literature analysis for both Genome Biology and BMC Bioinformatics journals are located in the data directory. In addition, a full list of references and links for the various major resources mentioned is located in the resource_bibliography directory.

The tool is finished and further described in our publication: Duck, G., Brass, A., Nenadic, G., Robertson, DL., and Stevens, R. bioNerDS: exploring bioinformatics' database and software use through literature mining. BMC Bioinformatics (2013) Paper Link

Thanks for your interest!

Corpora

Various gold standard corpora are available to browse and for download. Please cite my SMBM paper if you use these:

Duck, G., Stevens, R., Robertson, D., and Nenadic, G. Ambiguity and variability of database and software names in bioinformatics. Proc. of 5th International Symposium on Semantic Mining in Biomedicine (SMBM 2012), pages 2-9. Paper Link

Development

This tool has been developed in Java by Geraint Duck, under the supervision of Robert Stevens, David Robertson, and Goran Nenadic.

Program Installation and Execution (Standalone Example)

IMPORTANT WARNING: It has come to my attention that bioNerDS does not currently run under Java 1.8. I will endeavour to fix this when I get sufficient chance. In the mean time, I'm afaid you will have to run the program under an older install of Java.

1. Download the bioNerDS binary, the GATE plugins directory, the GATE JAPE files, the main software and database dictionary, the English dictionary, the acronym dictionary and the example properties file.
2. Extract both zip files (gate_plugins.zip and japes.zip).
3. Edit the properties.conf file if required (e.g., uncomment the noStanford line, etc.).
4. Input formats for processing can be specified through the command line using the same settings as LINNAEUS. For example:
   • --text <text file(s)>
   • --textDir <path> [--recursive]
   • --pmcDir <pmc directory> --dtd <dtds> [--recursive]
5. Execute on the command line with:

java -Xmx1200m -jar bioNerDS.jar --properties properties.conf --text <example_file.txt> 

1. Results will be tab separated and outputted in out.normal.txt by default (see .conf file).

License

All source code is released under the BSD 2-Clause License.

See license for full restrictions.

All included libraries retain their own respective licenses. Some source code from other projects has been used/adapted for inclusion in this project (e.g., LINNAEUS, JCommander, JOrtho, GATE and Snowball). Attribution for these remains with the original respective authors, and is distributed in accordance with their own licenses.