A New Method to Identify Non-Functional Olfactory Receptor Genes
Author Information
Author(s): Menashe Idan, Aloni Ronny, Lancet Doron
Primary Institution: The Weizmann Institute of Science
Hypothesis
Can a probabilistic algorithm accurately distinguish between functional and non-functional olfactory receptor genes?
Conclusion
The CORP algorithm can effectively identify non-functional olfactory receptor genes, predicting that about 70% of human olfactory receptor genes are likely non-functional pseudogenes.
Supporting Evidence
- The CORP algorithm identified only 5% of mouse functional olfactory receptors as pseudogenes.
- The algorithm predicted that 135 out of 384 human olfactory receptor genes are likely non-functional.
- The logistic regression model correctly classified 96.3% of intact mouse olfactory receptors.
Takeaway
Scientists created a computer program to tell which smell-related genes in humans don't work anymore, and they found that many more of these genes are broken than previously thought.
Methodology
The study developed a probabilistic algorithm called CORP that uses logistic regression to classify olfactory receptor genes based on their sequence conservation.
Potential Biases
Potential false positives may arise from the probabilistic nature of the conservation core and SIFT matrix.
Limitations
The algorithm may misclassify some recently-formed pseudogenes as functional due to insufficient deviation from conservation.
Participant Demographics
The study analyzed olfactory receptor genes from humans, mice, and dogs.
Statistical Information
P-Value
p<10-8
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website