Comparing two phylogenies

Question

[Jautis]

Hi,

I have two phylogenies (1 genetic, 1 phenotypic) that I would like to compare. I know that they are different phylogenies (that much I can tell by looking/with all.equal from ape) and can generate an overall similarity from the var-covar or distance matrix (using a mantel test).

What I would like to do now is separately determine how similar the topology and branch length are. Does anybody know of a way to do this?

Thanks!

[Andrzej Zielezinski]

Check my answer here: Identification Of Significant Differences Between Phylogentic Trees Using Distance Matrix

Answer 1

[Shicheng Guo]

1. quantitative: Shared branches / total branches
2. visual: circular graph

Answer 2

[jhc]

You may want to take a look at ete-compare. It allows one to compute unweighted robinson-foulds distances, as well as the number of partitions/branches in one tree that are not found in the other. It permits also to dump the matches and mismatches between trees.

Answer 3

[Brice Sarver]

There are a variety of metrics that compute distances between two trees. The most commonly used is the Robinson-Foulds (RF) distance, which looks for nodes present in one tree and vice versa. There are also methods that take branch lengths into account directly, like the Kuhner-Felsenstein Distance. These methods are implemented in R in several packages. Be aware that you will want your trees to have branch lengths in the same units if you are going to use any method that takes them into account.