BUSTED-E: A New Method for Detecting Natural Selection in Sequence Alignments
Author Information
Author(s): Selberg Avery G, Chikina Maria, Sackton Tim, Muse Spencer V, Lucaci Alexander G, Weaver Steven, Nekrutenko Anton, Clark Nathan, Pond Sergei L Kosakovsky
Primary Institution: Temple University
Hypothesis
Can BUSTED-E effectively reduce false positives in detecting positive selection due to alignment errors?
Conclusion
BUSTED-E significantly reduces the rate of false positives in detecting positive selection by accounting for alignment errors.
Supporting Evidence
- BUSTED-E reduces unrealistic rates of positive selection detection by an order of magnitude.
- BUSTED-E can annotate alignments with locations of possible errors.
- BUSTED-E improves biological interpretability of results.
- BUSTED-E is a scalable solution for evolutionary analyses.
Takeaway
BUSTED-E is like a smart filter that helps scientists find real signs of evolution in DNA sequences by ignoring mistakes in the data.
Methodology
BUSTED-E is a random-effects model that tests for positive selection while accounting for alignment errors.
Potential Biases
Potential for misinterpretation of biological signals as errors due to the heuristic nature of the error component.
Limitations
The method may not detect all real errors and could misclassify some biological variations as errors.
Statistical Information
P-Value
p<0.001
Statistical Significance
p<0.001
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website