Re: Digest for figtree-discuss@googlegroups.com - 1 Message in 1 Topic

[Brian Foley]

In my opinion phylogenetic trees should always use branch lengths that are meaningful, based on evolutionary distances.  The only reason I can think of for using the transform branches option, to create a cladogram, is if the data used to build the tree was not based on characters with well behaved evolutionary rates.  For example if you build a tree based on leaf size and other characteristics of plants, rather than on DNA or protein sequence data. Cladograms can give very misleading impressions of how evolutionary history has shaped groups of organisms.  In my opinion, the biggest choices to make in building good trees are 1) which data to include and what to use for an outgroup for the tree; and 2) which methods and choices of models of evolution to use, but these choices are trivial in comparison to choosing the right data to begin with. Brian Foley, PhD HIV Databases --- On Thu, 6/17/10, figtree-dis...@googlegroups.com <figtree-dis...@googlegroups.com> wrote: From: figtree-dis...@googlegroups.com <figtree-dis...@googlegroups.com> Subject: Digest for figtree...@googlegroups.com - 1 Message in 1 Topic To: "Digest Recipients" <figtree-dis...@googlegroups.com> Date: Thursday, June 17, 2010, 6:04 AM   Today's Topic Summary Group: http://groups.google.com/group/figtree-discuss/topics FigTree 1.3.1 phylograms [1 Update]  Topic: FigTree 1.3.1 phylograms lisa <lisa...@gmail.com> Jun 16 07:53AM -0700 ^   Hi, I am a student working on preparing trees for publication. I am new to the program and am in need of some help in preparing a phylogram. I see under Trees-Transform branches, there are three options: equal, cladogram, or proportional. Is the proportional option the option I am looking for, or does proportional mean it is proportional to the amount of change?   Thanks for any help.   -Lisa