Extracting Protein Interaction Relations from Biomedical Text
Author Information
Author(s): William A Baumgartner Jr, Zhiyong Lu, Helen L Johnson, J Gregory Caporaso, Jesse Paquette, Anna Lindemann, Elizabeth K White, Olga Medvedeva, K Bretonnel Cohen, Lawrence Hunter
Primary Institution: Center for Computational Pharmacology, University of Colorado School of Medicine
Hypothesis
Can concept recognition improve the extraction of protein interaction relations from biomedical literature?
Conclusion
Current information extraction technologies are nearing the performance levels needed for concept recognition to provide high-quality data to biologists.
Supporting Evidence
- The study demonstrated the effectiveness of concept recognition in biomedical language processing.
- It highlighted the challenges of ambiguity in gene and protein names.
- The system was built using a modular approach, allowing for quick experimentation with different methods.
- Results indicated that combining outputs from multiple gene taggers improved precision and recall.
Takeaway
This study shows how recognizing concepts in text can help scientists find important information about proteins and their interactions more easily.
Methodology
The study used a modular approach to build a protein interaction relation extraction system, integrating various tools and techniques for concept recognition.
Potential Biases
There is a risk of bias due to the reliance on specific training datasets that may not represent the broader literature.
Limitations
The performance of the system was affected by over-training on previous datasets and the ambiguity in gene and protein names.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website