Lecture 1

I converted my PowerPoint slides to a Quicktime movie that you can download and play to review. This should be possible on both Macs and PC's.

The assignment is to enter and analyze the Laurin and Reisz matrix. First, see if you can duplicate his results. Second, constrain temnospondyls and Lissamphibia to be a clade and analyze the data set under that constraint. Third, get down and funky and do something interesting with the data matrix.

I assigned groups as follows (we will switch in the future):

Greg and Holly
Rafe and Amy
Emily and Jonathan
Cat and Jackson
Alisha and Gabe
Ted and Nina

I'd like each group to present their results on the 23rd. Two overheads maximum, no more than about 5 minutes talking so that we can discuss this adequately. Don't re-hash the general stuff about the matrix.

I paired folks up so that those who know how to run PAUP can help the others, but it's my intent that everyone learn how to use it. So don't totally depend on your partner to do everything for you.

Some notes on running PAUP:

You can run PAUP using either the menus or commands embedded in the PAUP file. I like the latter way because it gives running record of what I've done. Knowing which commands to use can be daunting, but here are some tips.

  1. PAUP and MacClade data files are just plain old ASCII text files. They can be edited using MS Word, the PAUP editor, BBEdit, etc.
  2. The PAUP data file is structured into blocks; there is a TAXA block, a CHARACTERS block, etc. The commands to analyze the matrix are in a PAUP block. A block is a chunk of text that starts with "BEGIN keyword" and ends with "END;" For example:

begin paup;
hsearch;
showtree all;
end;

 

This PAUP block does a heuristic search, and then displays all of the trees found in that search. There are many implicit options for which the default value is used. We can change these as in the example below. Note that comments can be embedded anywhere in the datafile using brackets []. Also the datafile is in free format, meaning that the position of linebreaks and the number of spaces between commands does not matter.

begin paup; [begin the block; note the semicolon]
log file=dataset1.log append=yes; [log all the output into a text file named dataset1.log, but append this output to the file if it already exists, so that I don't overwrite it by mistake]
exclude 3-5 8 10 45-67;  [this command excludes certain characters from the analysis; the opposite command is include]
[you can put comments almost anywhere]  
constraints AmphibianMonophyly = (1-9,13,(10,11,12)); [define a constraint tree in which taxa 10,11,12 must be a clade]
constraints FrogMonophyly = (1-10,13,(11,12)); [define another constraint tree; you can do as many as you need]
delete 1-4 5 6; [delete the specified taxa from the analysis; the opposite command is restore;]
hsearch; [do a heuristic search, but note that the constraint is not enforced]
showtree 1; [show only the first tree from this search;]
include all; [re-include all of the characters]
hsearch swap=tbr [do another analysis with the following subcommands: use TBR branch swapping; note there is no semicolon]

constraints = AmphibianMonophyly

[this time, enforce this particular constraint tree; note again, no semicolon]

enforce=yes

[ no semicolon]
addseq=random nreps=100;
[to get a more through search of possible trees, select the next taxon to be added to the tree at random; then repeat the search 100 times in hopes of finding a shorter tree; note the semicolon, which ends the subcommands of HSEARCH]
savetrees /file=dataset1.trees; [save all the shortest trees found in a file named dataset1.trees; this file will contain the tree description in parenthetical notation; you can re-use these trees later for printing, etc.]
log stop; [stop logging to the file]
end;  
[!Because this comment starts with a !, it will be echoed (written) to the log file]  
   
   
  1. After you add or edit a PAUP block to your datafile, you must save the file (apple-S) and then execute the datafile (apple-R for RUN). Executing the datafile will check the syntax of the commands first, and alert you if there are errors. If there are no errors, the program executes the PAUP block. If you have just created a datafile but have no PAUP block, executing the datafile will just check the data and then terminate normally. This is useful for checking syntax errors in the data.
  2. You can keep your old PAUP blocks in the datafile by just adding [ ] around the entire block, which converts it to a comment; this is handy. You can also define several PAUP blocks in a row, each with different analyses, log files, etc. and then run all in one session, like overnight.
  3. You can use a SETS block to define sets of characters or taxa. Then, you can use the name of the set in include/exclude and delete/restore statements rather than having to list all of the numbers.

    BEGIN SETS;
    Charset Skull = 1-50;
    Charset Postcranial = 51-75;
    Taxset Lissamphibia = 23-29;
    Taxset Lepospondyls = 31-45;
    END;

  4. After you start PAUP you should see a window that has "PAUP* 4.0b8" or something similar as the title. At the bottom of this window you should see a one-line blank rectangle. This is the command-line box, and you can enter PAUP commands here. One very useful command is "?", a question mark, which will give you a list of the possible PAUP commands. If you enter a command followed by a ? then PAUP will give you the syntax of the command. This is useful in setting up PAUP blocks.
    You can also execute PAUP commands by typing them in here, rather than putting them in a PAUP block or using the menu.

 


Last modified 1/20/2002 by DCC