How to draw a phylogenetic tree with PHYLIP program for tumor samples

Question

[jennyp0706]

Hi, I'm dealing with WES samples consist of primary tumor, metastases(lymph nodes), normal from one patient.

I want to construct a phylogenetic tree using somatic SNVs. A binary data is used with normal sample as outgroup and for all somatic mutations found in the patient, if that specific somatic mutation exists in each metastases, it would be stated as 1(presense) and if it doesn't exists, it would be stated as 0(absense). My ultimate goal is to figure out which lymph node metastases first.

I notice that the PHYLIP program offers diverse methods. I'm thinking of using either 'pars' or 'penny' as the method but couldn't choose which one I should use. I personally think the assumption that penny uses (Changes 0 --> 1 are much more probable than changes 1 --> 0) is more consistent with the generally accepted theory of clonal progression in tumors. However, when I look through research articles that used PHYLIP, pars(Changes to all other states are equally probable (Wagner)) is way more commonly used and articles with 'penny' is hard to find.

Therefore, I am asking for help regarding this issue. Which method should I use for my tumor samples to construct a phylogenetic tree.

[GenoMax]

I think what you are looking for is to infer clonal architecture in your data using a program like SciClone, not a traditional phylogentic program like Phylip. See this paper for more information: https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1003665

Answer 1

[brian.fristensky]

The BIRCH system implements many of the Phylip programs in a convenient point and click interface. This includes a number of programs such as Dollo, Wagner and Pars for discrete state (0,1) data (Penny is available in BIRCH, but would have to be run at the command line). The tutorial PHYLOGENETIC ANALYSIS OF DISCRETE STATE DATA walks through an example of constructing trees, and the VISUALIZATION OF PHYLOGENETIC TREES tutorial shows how to create publication quality trees using Archaeopteryx.