BUCKy 1.2 README For higher versions, please refer to the user's manual. INSTALLATION Pick a directory where you want the BUCKy code to be. I will call this directory $BUCKY_HOME in this documentation. To open the compressed tar file with the BUCKy source code and example data, do these commands. cd $BUCKY_HOME tar zxf bucky-1.2.tgz This creates a directory named BUCKy-1.2 with subdirectories BUCKy-1.2/data and BUCKy-1.2/src . COMPILATION If you have gcc installed, compile the software with these commands. cd $BUCKY_HOME/BUCKy-1.2/src make This will compile programs mbsum and bucky. I suggest putting copies in ~/bin if this is in your path. If you do not have gcc installed, you need to find the installer. On a Macintosh, it may be in Applications/Installers/Developer Tools . HELP Type these commands for very brief help messages. mbsum --help bucky --help EXAMPLE Suppose that you have a directory where each file is of the form *.t and is a MrBayes output file. Use mbsum to summarize each file. Remove the first 1000 trees of each for burnin. for X in *.t; do mbsum -n 1000 $X; done This will create a file named <filename>.in for each file named <filename>.t . Warning! It will overwrite files with the name <filename>.in if they exist. Next, to run bucky with default parameters, try this. bucky *.in This will create a bunch of output files of the form run1.* . You can pick your own root file name. YEAST EXAMPLE To try the yeast example described in the Ane et-al MBE paper (with a much smaller number of updates), you can try the following. cd $BUCKY_HOME/BUCKy-1.2/data/yeast ../../src/bucky y???/*.in EXAMINING OUTPUT Mac users who want to use standard Mac applications to read the output can open any of the output files using TextEdit. Bret Larget and Cecile Ane 17 January, 2007 Madison, Wisconsin