ART is an ancestral reconstruction tool . . .

NAVIGATE: overview | getting started | installation | contents | command-line syntax | projects | ART output

Getting Started

(For new users)

Let's follow a sample use-case . . .

0. Install ART

If you haven't already done so, follow the installation instructions.

1. Prepare your input data

At its core, ART is a CodeML wrapper. CodeML requires three input files: a set of hypothesized phylogenetic trees, a set of taxon sequences, and the name of an evolutionary model. Suppose we have these three files, located at:

     ~me/myfiles/trees.txt
     ~me/myfiles/sequences.txt
     ~me/myfiles/modelname.txt

2. Run CodeML

First, we need to run codeML on our input data. Let's call our new project, 'myproject'.

At a command prompt, type:

art calculate -p myproject -treefile ~me/myfiles/trees.txt -seqfile ~me/myfiles/sequences.txt -modelfile ~me/myfiles/modelname.txt

TIP: If you recieve the following error message: 

   art: command not found

Instead, try:

  $ARTHOME/art calculate -p myproject . . . .

CodeML will begin processing the data, and output will appear on the screen. Be patient, this step might take several minutes (or longer).

When CodeML is finished, your MySQL database will contain new data. The new tables will have 'myproject' in their name (where 'myproject' is the name of your project). Feel free to look in MySQL and verify this is true.

You will also notice a new directory, named '$ART/myproject/'. Inside this directory, you will find copies of your input files, along with some directories for output and temporary files.

3. Find ancestors

Now that our database is filled, we can use ART to search for ancestors. Suppose we want to find the ancestors for taxons #13 through 35. Also suppose we want to use the tree where taxa #47 through #53 are the outgroup. Finally, we also want to find the maximum a posteriori sequence.

At a command prompt, type:

art findanc 13-35 -p myproject -outgroup 47-53 -map

This will place several files into the directory $ART/myproject/output. . .

tree_node_pairs.txt

is a table of tree IDs and corresponding ancestor IDs.

reconstructed_sequences.txt

is a table of tree IDs and ancestral sequences.

sequence_table.txt

is a table of ancestral sites and states for each tree.

map_ancestral_sequence.txt

contains the MAP ancestral sequence, with detailed site-by-site information. This file will only appear if you use the -map argument with the findanc command.

The 'findanc' command can also be customized with many other possibilities. Click here for more information about the ART commands.

4. Find tree information

Next, suppose we want to review information about our hypothesized phylogentic trees.

At a command prompt, type:

art showtrees -p myproject

After this command is complete, one new file will appear in '$ART/myproject/output':

tree_info.txt

contains a table with tree IDs, the tree (in parenthetical notation), and the tree's probaility.

5. Delete a project

If you decide to remove a project, ART can save you the trouble of deleting files and manually removing MySQL tables. Suppose we want to remove 'myproject'. . .

At a command prompt, type:

art clean -p myproject

** A note on shell scripts **

ART works well with shell scripts, especially if you want to run several commands or if you are dealing with multiple datasets. To get you started, the ART package contains a sample shell script, named 'runall.sh' This script is located in the top-level ART directory. Depending on your shell configuration, this script should run with one of the following commands:

    './runall.sh'
    'source ./runall.sh
    'bash ./runall.sh

Please feel free to experiment with 'runall.sh' and modify it for your own purpose.