How To Test If A Phylogenetic Tree Matches A Topological Hypothesis?

Question

[qiyunzhu]

Dear all,

I am trying to see if the phylogenetic trees I constructed support a topological hypothesis that a certain group of organisms is monophyletic. I can of course eye-ball the tree to see if it's correct, but the question is that, I have thousands of trees... Actually I am trying to compute the posterior probability of a topological hypothesis from a MrBayes tree sample file (.t). Therefore, I am seeking an automatical way to scan all these 1000 trees and count how many of them match a topological constraint I provide. The principle is simple. I'm just wondering if there is already a tool out there. Anyone has any thoughts? Thanks!

Best wishes

[Whetting]

look into hypothesis testing. The KH test is pretty good and is implemeted in many tools. E.g. check this presentation by Mark Holder http://phylo.bio.ku.edu/woodshole/HolderTestingLectureWoodsHole2012.pdf

Answer 1

[Csaba Kerepesi]

Hi!

There is a perl script for calculating monophyletic values of gene families (for bacteria and archaea).

See monophyly value section of this article: http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0077033

Scripts are available here: http://edhar.genomecenter.ucdavis.edu/~dwu/BAmarker/ monophyly_test.pl in scripts.tgz)

I hope it helps.

Answer 2

[Dave Lunt]

You should also have a look at ETE, which has test for group monophyly built in and would be perfect for iterating across lots of trees: "ETE is a python programming toolkit that assists in the automated manipulation, analysis and visualization of phylogenetic trees" https://pythonhosted.org/ete2/index.html

Answer 3

[yukichanyk]

Have you tried topological hypothesis testing in MrBayes with Bayes factor comparisons? Check the mb3.2 manual chapter4.4 :)