Predicting Genetic Interactions in Yeast Using Protein Networks
Author Information
Author(s): Paladugu Sri R, Zhao Shan, Ray Animesh, Raval Alpan
Primary Institution: Keck Graduate Institute of Applied Life Sciences
Hypothesis
Can the network properties of proteins predict their joint dispensability in synthetic genetic interactions?
Conclusion
Protein interaction networks can effectively predict synthetic lethal interactions with high accuracy.
Supporting Evidence
- The method achieved sensitivity and specificity exceeding 85%.
- The prediction performance was robust against errors in the protein interaction network.
- The study identified novel synthetic sick/lethal gene pairs at a genome-wide scale.
Takeaway
Scientists used a computer program to look at how proteins interact in yeast to guess which pairs of genes might work together to cause problems when one is missing.
Methodology
A support vector machine was trained using graph-theoretic properties of proteins to predict synthetic sick/lethal interactions.
Potential Biases
Potential bias due to reliance on curated data and the exclusion of certain gene pairs.
Limitations
The method relies on high-quality protein interaction data, which may not be available for all organisms.
Participant Demographics
The study focused on Saccharomyces cerevisiae, a type of yeast.
Statistical Information
P-Value
< 2.2 × 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