Text Detective: A System for Gene Annotation in Biomedical Texts
Author Information
Author(s): Tamames Javier, Christian Blaschke, Lynette Hirschman, Alfonso Valencia, Alexander Yeh
Primary Institution: Alma Bioinformatics S.L.
Hypothesis
Can a rule-based system effectively identify and normalize gene mentions in biomedical texts?
Conclusion
Text Detective achieves high precision and recall in identifying gene mentions in biomedical literature.
Supporting Evidence
- Text Detective achieved 84% precision and 71% recall for gene identification.
- The system is capable of annotating a wide range of biological entities.
- Performance improves with longer texts due to more context information.
Takeaway
Text Detective is like a smart helper that finds and names genes in science articles, making it easier for researchers to understand them.
Methodology
The system uses a combination of rules and biological lexicons to identify and normalize gene mentions.
Potential Biases
Errors may concentrate in difficult cases where gene names are common acronyms.
Limitations
The system's performance can vary based on text length and the ambiguity of gene names.
Statistical Information
P-Value
p<0.05
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website