FIRE-pro tutorial

Installing and using FIRE-pro: a basic tutorial

Download

FIREpro can be downloaded and installed on Unix, Mac, or PC (via Cygwin).

The package comes pre-loaded with data files for S. cerevisiae. For any other species, follow the instructions entitled "Species data".

For sample expression files for each species, download this package.

Installation

1) Unzip the .zip file using unzip (files will be unzipped in a FIREpro-1.1/ directory)

unzip FIREpro-1.1.zip

Then, enter the directory and run the make tool:

cd FIREpro-1.1/ make

If you ever need to recompile the program in the future, make clean undoes make and vice versa.
Examine the output of make to make sure that were no compilation errors (ignore warnings).

2) Create a $FIREPRODIR environment variable, that points to where FIREpro is installed (i.e. the current directory). On many Unix platforms, the command will be something like:

export FIREPRODIR=/path/to/FIREpro-1.1

The /path/to/ should be replaced with the actual path. If you are currently in the FIRE-pro-1.1 folder, a simple way to set FIREPRODIR is to use `pwd` as in:

export FIREPRODIR=`pwd`

For longer term export of the $FIREPRODIR variable, open the shell initialization file (e.g., ~/.bashrc) and add the command: export FIREPRODIR=/path/to/FIREpro-1.1. If export is not present, it can be replaced by setenv as in: setenv FIREPRODIR /path/to/FIREpro-1.1 or setenv FIREPRODIR `pwd`

3) The sequence files for the budding yeast S. cerevisiae have come pre-packaged in FIREpro-1.1.zip. To analyze data from other species, one can download the pre-packaged sequence files for your organism(s) as detailed in the section "Species data".

4) To view the output of FIRE-pro as pdf and png images, the commands ps2pdf and convert, must be available. These commands are part of the Ghostscript and ImageMagick packages, respectively. Most Unix servers already have these installed, whereas a local machine such as a PC running Cygwin or a Mac running Terminal may not have these by default, but they can be easily installed. To check if these packages are already installed, enter the bash prompt and type ps2pdf or convert. If the shell does not recognize the command, the program is not installed or the proper directory has not been added to the $PATH variable using export or setenv. For Mac installation of these programs, use MacPorts as detailed here. For installation on PC, these packages can be downloaded and installed via Cygwin. For Unix or Linux, machines these packages can be individually downloaded at Ghostscript and ImageMagick. Even without ps2pdf and convert, FIRE-pro should still run properly and will be export images as .eps files, which can be converted to .pdf or .png formats independently.

Species data

The sequence files for the budding yeast S. cerevisiae have come pre-packaged in FIREpro-1.1.zip. To analyze data from other species, one can download the pre-packaged sequence files for your organism(s). Download the pre-packaged sequence file(s) for your organism(s) to your FIREpro-1.1/ directory, then simply unzip the file with Unix:

Currently supported organisms are:

Or click here to download data for all available species.

For example, the following downloads and installs the files for mouse:

cd $FIREPRODIR wget https://tavazoielab.c2b2.columbia.edu/FIRE-pro/species_data/mouse.zip unzip mouse.zip rm mouse.zip

The above commands will create a FIREpro-1.1/Data/ directory with the proper file hierarchy for each species. Of course, you don't need to download all of the species files above, just the ones you want to work with. Assuming everything went smoothly, FIRE-pro can now be used. Some sample protein profiles can be found here.

Running FIRE-pro

The current implementation of FIREpro is best executed from the main directory (FIREpro-1.1/). The basic command line syntax for FIREpro is:

perl $FIREPRODIR/fire_aa_wrapper.pl --expfile=<inp> --exptype=<type> --species=<sp> --runname=<dir/prefix>

where <inp> indicates the input expression profile (e.g., "exp.txt", <sp> indicates the species (e.g., "yeast"), and <type> refers to the expression profile and can be discrete (e.g., cluster indices) or continuous (e.g., half-life values).

A sample command line is below:

perl $FIREPRODIR/fire_aa_wrapper.pl --expfile=expfiles/YBR160W.txt --exptype=discrete --species=yeast --kmers=3-5 --gaps=0-2 --runname=Out/ybr160w_interaction/ybr160w

Species name can be chosen from: yeast, worm, fly, human, mouse, spombe, plasmodium. The corresponding pre-packaged sequence files must be downloaded (see "Species data").

A FIRE-pro analysis should take ~1 hour on a Mac or Unix server.

Example

Here is a sample workflow that can be applied directly to the existing FIRE-pro package after download:

unzip FIREpro-1.1/ cd FIREpro-1.1/ export FIREPRODIR=`pwd` make perl $FIREPRODIR/fire_aa_wrapper.pl --expfile=expfiles/YBR160W.txt --species=yeast --exptype=discrete --runname=Out/ybr160w_interaction/ybr160w perl $FIREPRODIR/fire_aa_wrapper.pl --expfile=expfiles/half_life_belle.txt --species=yeast --exptype=continuous --runname=Out/half_life/half_life perl $FIREPRODIR/fire_aa_wrapper.pl --expfile=expfiles/oshea_mitochondrion.txt --species=yeast --exptype=discrete --runname=Out/mito_localization/mito