Mining the Caenorhabditis Genetic Center Bibliography for Genes Related to Life Span
Author Information
Author(s): Blei DM, Franks K, Jordan MI, Mian IS
Primary Institution: Princeton University
Hypothesis
Can statistical modeling of biomedical corpora reveal insights into genes associated with life span?
Conclusion
The study demonstrates that LDA can extract meaningful insights and generate predictions about genes related to life span modification.
Supporting Evidence
- LDA outperformed simpler models in predicting document relevance.
- The model identified novel candidates for age-related genes.
- Topics generated by LDA captured meaningful aspects of nematode biology.
Takeaway
Researchers used a computer model to look at a big collection of scientific papers to find genes that might affect how long worms live.
Methodology
The study used the Latent Dirichlet Allocation (LDA) model to analyze 5,225 documents from the Caenorhabditis Genetic Center Bibliography.
Potential Biases
The curation of the CGC Bibliography may introduce bias in the selection of genes discussed.
Limitations
The model is based on a simple bag of words representation, which may overlook syntax and semantics.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website