Link to the ExPASy Molecular Biology server of the SIB

Link to official BioMinT home page

          Link to corpora produced by SIB

Abstract

Genome research has spawned unprecedented volumes of data, but characterisation of DNA and protein sequences has not kept pace with the rate of data acquisition. To anyone trying to know more about a given sequence, the worldwide collection of abstract and papers remains the ultimate information source. The goal of the BioMinT project is to develop a generic text mining tool that:

• interprets different types of queries

• retrieves relevant documents from the biological literature

• extracts the required information

• outputs the result as a database slot filler or as a structured report

The tool will thus provide two essential research support services: (1) Curator's assistant: accelerate, by partially automating, the annotation and update of bio-databases; and (2) Researcher's assistant: generate readable reports in response to queries from biological researchers.

The Swiss Institute of Bioinformatics as partner of the BioMinT project

As user of the annotation assistant prototype that will be developed, the SIB will is in particular involved in:

• defining program specifications: it will work actively on user requirement analyses, data corpus production, and domain-specific knowledge collection

• furnishing sets of document corpora for each stage of tools development and validation

• exploring domain-specific knowledge resources that could be exploited in the development

• testing the preliminary prototype: evaluation will focus on the performance (precision and recall) the IR/IE modules as well as on the efficiency and usability of the overall system. Users will perform test runs and produce critical feedback in view of prototype revision and improvements for integration into the final prototype

• providing user requirements, corpus and validation test for the update module and implement a module for systematic update annotation of Swiss-Prot database entries

• validation phases/test runs of the generic prototype, especially in the final validation test by providing direct feedback from database curators.

SIB will exploit the resulting BioMinT prototype internally. This tool will add value to a protein sequence by including descriptive information extracted from the scientific literature as well as from other web knowledge sources in their Swiss-Prot database. In particular, BioMinT is expected to: increase the relevance and completeness of added information to speed up the annotation procedure in order to handle the amount of new incoming sequences.

For members only

Project partners

Corpora produced by SIB