Estimating Covariance of Evolutionary Distances from Pairwise Alignments
Author Information
Author(s): Dessimoz Christophe, Gil Manuel
Primary Institution: ETH Zurich
Hypothesis
Can we develop a covariance estimator for distances estimated from pairwise alignments of biological sequences?
Conclusion
The new covariance estimator performs comparably to the maximum likelihood variance estimator, with biases that are manageable for distances below 150 PAM units.
Supporting Evidence
- The covariance estimator can be used with the maximum likelihood variance estimator to form covariance matrices.
- The estimator shows no bias for sequence divergences below 150 PAM units.
- The study involved 40,000 Monte Carlo simulations to validate the estimator's performance.
Takeaway
This study created a new way to measure how similar two sequences are without needing to align all sequences at once, which is usually very hard to do.
Methodology
The study used Monte Carlo simulations to analyze the performance of the covariance estimator.
Potential Biases
The estimator is biased when the percentage of correctly aligned positions (anchors) is low.
Limitations
The estimator may underestimate covariances at high sequence divergences.
Statistical Information
Confidence Interval
95%
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website