There is a widespread need for effective literature mining to help biologists gather and make use of information dispersed in multiple text documents. We have developed a system that investigates the use of text data mining techniques to build computer model networks of biochemical pathways. The system uses gene lists and text articles specified by the user. These are processed by parsing the text and detecting element associations to reveal biological information related to interactions between gene products. It produces graphical representations in real time to enhance visualisation. Graphical outputs can also be presented textually as a System Biology Markup Language (SBML), a new method for representation and exchange of biochemical models. These insights, methods and systems have wide application to understanding the patterns and processes of gene product relationships.
Last modified: Thursday, 14-Oct-2004 14:37:53 NZDT
This page is maintained by Caroline Wills.