Related Experiment Videos
Unrooted genealogical tree probabilities in the infinitely-many-sites model
1Department of Mathematics, Monash University, Clayton, Australia.
Mathematical Biosciences
|May 1, 1995
Summary
The infinitely-many-sites process models DNA sequence variability but its sampling theory is unclear. This study clarifies its tree structure, enabling probability calculations and estimators for evolutionary rates.
Area of Science:
- Population genetics
- Computational biology
- Evolutionary modeling
Background:
- The infinitely-many-sites model is widely used for DNA sequence variability.
- Its underlying sampling theory remains poorly understood, limiting its application.
Purpose of the Study:
- To elucidate the sampling theory of the infinitely-many-sites process.
- To develop computational methods for estimating evolutionary parameters.
Main Methods:
- Describing the tree structure inherent to the model.
- Deriving probability recursions for sequence configurations (trees).
- Implementing a Monte Carlo recursion for approximate sampling probabilities.
Main Results:
- Demonstrated how to compute sample sequence probabilities using the tree structure.
- Developed methods to infer unrooted and rooted genealogies from site data.
- Provided a computational algorithm for approximate sampling probabilities for any sample size.
Conclusions:
- The study clarifies the sampling theory of the infinitely-many-sites process.
- The developed methods facilitate the computation of probabilities and estimation of substitution rates.
- The algorithm aids in both known and unknown ancestral labeling scenarios.