TABASCO

From OpenWetWare

Revision as of 17:42, 6 November 2007 by Endy (Talk | contribs)
Jump to: navigation, search

Home        Lab Members        Research        Notebooks        Publications        Internal        Contact       


Contents

Summary

TABASCO is a simulator created to address the problem of simulating gene expression at single-base resolution. By defining the logic of transcription and translation rules a priori such as initiation, elongation, termination, and interactions of polymerases and proteins, Tabasco automatically traverses the state of the system as it develops and thus makes simulation at such high resolution computationally feasible. Tabasco was designed to allow us to better understand bacteriophage gene expression. In general, Tabasco would be useful to those interested in explicitly simulating hypotheses of protein-DNA interactions and their relation to gene expression (e.g., eukaryotic gene expression initiation).

Note regarding trademarks

Our use of the word TABASCO here refers solely to a simulator of gene expression and other systems comprised of elementary chemical reaction events that can be ordered along one or more spatial dimensions. The origins of our use of the word were as an acronym, abbreviating the words "Transcription And Binding And Serious Computational Overhead." Many other uses of the word "Tabasco" are held as registered marks of the McIlhenny Company, purveyors of famous pepper sauces and other fine products.

People

Sriram Kosuri, Jason Kelly, Drew Endy

Methods

Tabasco avoids previous ‘combinatorial explosion’ problems by tracking the position and state of proteins and genetic elements on the DNA to dynamically generate appropriate reactions, such as promoters being blocked by traversing polymerases. In order to improve computational efficiency, Tabasco makes use of a Gibson-accelerated Gillespie SSA to compute the reaction event timing and the resultant time-evolution of the genetic system.

Software and Installation

Software

The software for Tabasco is freely available here:
Extractable archive of all code, documentation, and examples
Java Source Files
Javadoc Documentation
Compiled Java Byte Code

The simulator was only tested on J2SE version 1.4.2 and higher. A basic knowledge of running Java programs is needed. Generally one can either download the source, and then compile it using a java compiler to create byte code or download the precompiled byte code above.

Installation instructions (for Unix systems)

  1. Download the extractable archive.
  2. Extract the archive using the following command:
     tar -xvzf tabasco.tar.gz 
    • Windows users should be able to use WinZip or other free programs (or download Cygwin to make the process more unix like)
  3. Make sure java is in your path
    • To check, type
      java -version
      and make sure the command is found (also a good time to make sure you are above version 1.4.2). If you do not have an up-to-date version of Java, visit the download site and follow the installation instructions.

Usage

There are 3 main classes that can be run. TabascoSimulator.class actually executes the simulation and is the longest program. Averager.class averages output from multiple TabascoSimulator simulations. TabascoJpegMake is the visualizer that makes JPEG stacks to visualize data that can later be made into movies.

TabascoSimulator

The TabascoSimulator class executes simulations. The basic usage is as follows.

java TabascoSimulator inputfilename outputfilename [random seed]
  • The inputfilename is the location of the location of the input file.
  • The outputfilename is the prefix to be used for the output files.

The optional random seed can be used to override the input from the inputfilename. This field is useful if one is using a script to run simulations on a cluster.

A sample input file is included in the distrubution. To run it, follow this procedure:

  1. Go to the TabascoWeb/classes directory by typing
    cd TabascoWeb/classes
  2. Try to run a sample program
    java TabascoSimulator ../examples/t7_input_file.txt ../output-test- 

Averager

The Averager class averages output from the TabascoSimulator. The basic usage is as follows.

java Averager #_of_iterations outputfilename
  • The #_of_iterations is an integer that specifies the number of simulations you want to average together
  • The outputfilename is the prefix to be used for the output files.

The example simulation run in TabascoSimulator can be averaged as follows:

java Averager 2 ../output-test-

TabascoJpegMake

This class makes a set of jpeg's from the output files to visualize the DNA and molecule levels as a function of time.

java TabascoJpegMake output-filename mol-input-filename dna-file1 dna file2 ...
  • output-filename is the filename prefix for jpeg's that are output
  • mol-input-filename is the Molecule file name that you want to visualize
  • dna-file1 and dna-file2, etc are the DNA files that will also be a part of the visualization.

For the example input, you can visualize the output by running the following command.

java TabascoJpegMake ../vis- ../output_AVG.txt ../output-test-DNA_phage1_sim1.txt 
../output-test-DNA_phage2_sim1.txt ../output-test-DNA_phage3_sim1.txt

Examples

QuickTime movie showing TABASCO being used to simulate gene expression for the first 1500 seconds of bacteriophage T7 development.

Personal tools