Email updates

Keep up to date with the latest news and content from BMC Bioinformatics and BioMed Central.

Open Access Methodology article

A graph-search framework for associating gene identifiers with documents

William W Cohen123* and Einat Minkov2

Author Affiliations

1 Department of Machine Learning, Carnegie Mellon University, 5000 Forbes Avenue, Pittsburgh, PA 15213, USA

2 Language Technology Institute, Carnegie Mellon University, 5000 Forbes Avenue, Pittsburgh, PA 15213, USA

3 Center for Bioimage Informatics, Carnegie Mellon University, 5000 Forbes Avenue, Pittsburgh, PA 15213, USA

For all author emails, please log on.

BMC Bioinformatics 2006, 7:440  doi:10.1186/1471-2105-7-440

Published: 10 October 2006

Additional files

Additional File 1:

The code used for graph-search and learning-to-rank in this paper has been submitted as an additional file: a gzip-compressed tar file, with all code under an open source license. The code requires installation of Minorthird, another open-source package. The README.txt file in the top-level directory details how to compile and use the code. The code for combining NER and soft matching is implemented a special case of the graph search that uses a restricted set of paths through the graph, realized using the class SoftDictEntitySearcher. Code availability and requirements All code is implemented in Java, and has been verified to run on both Windows XP (with cygwin) and Red Hat Linux environments. The most recent version of the code is available via anonymous CVS from the authors, using the following commands:

export CVS_RSH=ssh

export cvsroot=:pserver:anonymous@raff.ml.cmu.edu:/usr1/cvsroot

% cvs login

(anything as a password)

% cvs checkout ghirl

Format: GZ Size: 1009KB Download file

Open Data