Related Experiment Videos
Textmining in support of knowledge discovery for vaccine development.
Christian Schönbach1, Takeshi Nagashima, Akihiko Konagaya
1Biomedical Knowledge Discovery Team, RIKEN Genomic Sciences Center (GSC), Yokohama 230-0045, Japan. schoen@postman.riken.jp
Methods (San Diego, Calif.)
|November 16, 2004
Summary
Developing computational tools for vaccine development is crucial. A new system, Gene expression/epitopes/protein interaction (GEpi), aids understanding of immune response by analyzing host-pathogen interactions from literature data.
Area of Science:
- Bioinformatics
- Immunology
- Computational Biology
Background:
- Vaccine development relies on analyzing vast amounts of data and literature.
- Manual curation of this information is a significant bottleneck.
- Automated tools are needed to accelerate the interpretation of results and database maintenance.
Purpose of the Study:
- To develop a knowledge discovery support system (GEpi) for enhanced understanding of immune response and vaccine development.
- To focus on molecular functions of HIV-infected T-cells and HIV epitope information.
- To integrate biomolecular data from domain-specific databases with MEDLINE abstract information.
Main Methods:
- Utilized text mining and interrelation of biomolecular data.
- Developed the Gene expression/epitopes/protein interaction (GEpi) prototype system.
- Focused on extracting molecular interaction, disease associations, and Gene Ontology (GO) functional information.
Main Results:
- GEpi successfully generated intuitive knowledge reports aiding host-pathogen interaction interpretation.
- Extraction of molecular interaction, disease associations, and GO information proved effective.
- Epitope information in MEDLINE abstracts was sparse and lacked context (e.g., HLA-restriction).
Conclusions:
- Text mining can support vaccine development by extracting molecular and functional data.
- Epitope information is often found in full-text articles (tables, figures, legends), not just abstracts.
- Domain-specific extraction rules for full-text articles and a knowledge model for epitope data are needed for effective text mining in vaccine development.