Intro and Install | FAQ | Command Line


Concepts


The idea of threading is to use an existing protein structure to model the structure of a new amino acid sequence.  Prospect specializes in the creation of the alignment, and the recognition of the protein that will provide the best template for the modelling.  Once the protein template and sequence alignment are decided upon, that information can be turned over to a modeling program like Modeller.

Target files are typically in the format SSP or XML.  SSP stands for secondary structure prediction.  An XML file can contain an SSP file, inside the <ssp>...</ssp> section.  XML files can also contain IFA (indel frequency array information) in the <singleins>...</singleins> and <singledel>...</singledel> sections.

Prospect Searches the directories in the TEMPLATE_PATH environmental variable to find possible templates.  Template files take the form of <template_name>.xml.  Prospect will also search directories listed in the {PREFIX}/share/prospect/template_paths file.




Vocabulary


target (query) The protein for which you only know the amino acid sequence.
template The protein for which you know the structure
template library A set of templates that should cover the breadth of different types of protein structures that are available in the PDB database
integer programming A method to optimize the alignment using a combinatorial method that allows for pairwise energies to be taken into account.  Much slower, but more effective for distant alignments





Installation



Other Software/Databases You'll need


The Prospect default template database
The NCBI toolkit (for PsiBlast)
The NR database (file1, file2, file3) (for homologue searches to create the PSSM)
PsiPred (for secondary structure prediction)

flex - Feature equation parsing
libxml - So prospect can manipulate XML files
Subversion - Source code checkout (until we officially release an official version)

Packages that will help if you want to do more advanced things:


swig - For python compilation
A MPI implementation (OpenMPI, Lam, Mpich1, or Mpich2 ) - For parallel protein threading
COIN-OR - If you want to do Integer Programming Based threading
GSL - If you want to do some of the parameter optimization (not needed)
DSSP - Secondary structure definitions for template generation



To download and compile, try the following:


If you just want to run with dynamic programming based threading:


svn co https://openprospect.svn.sourceforge.net/svnroot/openprospect/trunk openprospect
cd openprospect
./configure --prefix=$HOME/prospect_install
make; make install

If you to do Integer Programming


svn co https://projects.coin-or.org/svn/Cbc/trunk Coin-Cbc
cd Coin-Cbc
./configure --prefix=$HOME/prospect_install
make; make install
cd ../openprospect
./configure --prefix=$HOME/prospect_install --with-coin=$HOME/prospect_install
make; make install