Gene Ontology term overlap as a measure of gene functional similarity
Author Information
Author(s): Meeta Mistry, Paul Pavlidis
Primary Institution: University of British Columbia
Hypothesis
Can term overlap serve as a reliable measure of gene functional similarity compared to other semantic similarity measures?
Conclusion
Term overlap can serve as a simple and fast alternative to other approaches that require complex pre-calculations.
Supporting Evidence
- Term overlap is highly correlated with other semantic similarity measures.
- Term overlap is faster to compute than information content-based measures.
- Term overlap avoids some problems that affect probability-based measures.
Takeaway
This study shows that comparing gene functions can be done quickly by looking at how many terms they share, which is easier than using more complicated methods.
Methodology
The study computed term overlap for randomly selected gene pairs from the mouse genome and compared it with other semantic similarity measures.
Potential Biases
Potential biases may arise from the shallow annotation problem, where genes with few annotations can yield misleading similarity scores.
Limitations
The study may not account for all complexities of gene functional relationships and relies on the quality of GO annotations.
Participant Demographics
The study focused on gene pairs from the mouse genome, specifically 100,000 pairs.
Statistical Information
P-Value
p<10-16
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website