I'm running a phylogeny in PAUP, and I'm looking for a way to get the "describe trees" function to operate on a majority rule consensus tree, instead of simply describing any one of the 500 individual trees the analysis generated. Is there a way to do this?
Optimizing apomorphies has nothing to do with branch length. Cladograms don't have branch lengths, either, but you can certainly map character state transformations on a single cladogram. The main problem here is that a majority rule consensus tree is a compromise of (in this case) 500 trees that may have alternative optimizations of a given character on different fundamental trees. As Samuel suggested, you could export your trees or the consensus tree to Mesquite and look at the optimizations of the characters you are interested in on those. Another approach would be to filter the characters of interest - characters that are parsimoniously explained by the data probably have an unambiguous optimization on all the trees in your consensus set (although that is an empirical question to test, rather than a certainty).
Hi,
this works on the majority rule tree for me:
After having analyzed the data in PAUP* save the resulting consensus tree in nexus format. Take the character matrix you used, and save it as a simplified nexus (for example in the program Mesquite under "file -> export -> simplified nexus"). Then open both in PAUP with "execute [datamatrix filename];" and then "gettrees file=[tree filename];". once these are in PAUP go to the File menu and select "Log Output To File" (this will print the apolist to a text file that you can then open in a text editor). It is important to turn this on before executing the rest of the next command. Then use the command "describetrees 1/apolist=yes". After that you will have a complete apomorphy list in your log file. PAUP* will assign a number to each resolved node and then list the apomoprhies.
... it's not very pretty but will give you the information. was that what you were looking for?
I suppose you can export the consensus tree out and run character optimization in Winclada, if that's what you want to do?
- Badges
- Science method