SH Test, Shimodaira Hasegawa test
Uses bootstrap procedure
Gives us range of trees with different possibilities, lets us refute
tests whether alt hypothesis is significantly different than best phylogeny
has to create distribution to test hypothesis, uses bootstrap
Drawbacks of SH test
same as bootstrap, a little overconfident in estimations
Likelihood ratio test LRT, this test is very flexible
1. is one phylogeny more significant than another
2. molecular clock - can test whether or not we can use that data for divergence time. divergence times, valid assumption to know if they're moving in a clocklike manner. in data thats evolving neutrally, only in limited situations
3. models of evolution - which one is the best one for our data set
4. is natural selection working on this gene. can test whether a gene or parts of gene are under selection / looking for signs of natural selection in protein coding genes.
How does one select a model of evolution?
Do a stepwise comparison of all the different models, which is statistically different and determine which one to choose. multiple tests to find best one
What is the common process used to do this and what are some of the considerations that must be addressed during this process?
Fo a variety of multiple tests, and look at how they are treating different aspects, systematic way to do it.