Timeframe

How long will it take to set up DIGS and run a screen?

Step 1: installing the required components for DIGS

In principle this step should be straightforward, but there is some uncertainty as it depends on your platform and operating system. All of the components of DIGS (PERL, BLAST, MySQL) and well-established and widely used within bioinformatics, and in theory should be straightforward to install. A typical bioinformatics server will usually have most if not all of these programs installed already. If installations are required, they should only take a few minutes, especially for experienced bioinformaticians working on LINUX/UNIX operating systems. Away from LINUX, things are less predicable. On Macintosh computers one PERL library (DBD::MySQL) does not come as standard, and may not work 'out-of-the-box'). For more information about installing DIGS on a Mac, please see here.

Step 2: setting up your DIGS screen

Setting up a screen means choosing probes, references and targets, and formatting these for use within DIGS. As much as you may want to get going with screening, it is vital to take some time here. At this point you will need to frame the question that you hope to answer through screening. Which target genomes and why? What kind of sequences are you looking for? Do you expect there to be cross-matching to other kinds of sequences that you're not interested in?

Step 3: creating a control file

This should only take a few minutes. See here for information about the control file structure.

Step 4: running the screen

Similarity search based screening is a computationally intensive procedure, and can take hours to days to complete. The DIGS output will show how far along the screen is in terms of queries executed, but bear in mind that the length of time each query takes will depend on the size of the target file and the scarcity/abundance of sequences matching the probe in the target file. Where screens are being run on a server, and are expected to take several hours or days to complete, you may want to run them in the background, as described here, so that your DIGS run is not dependent on maintaining a connection to the server.