The mcBPPS program version 1.0: June 3, 2010.  Written by:
   Andrew F. Neuwald
     The Institute for Genome Sciences and 
     the Department of Biochemistry & Molecular Biology,
     University of Maryland School of Medicine, 801 West Baltimore St.
     BioPark II, Room 617, Baltimore, MD 21201

Uncompress mcbpps.tar.gz by typing "gunzip mcbpps1_0.tar.gz"
untar the file by typing "tar xvf mcbpps1_0.tar"

This creates a directory 'mcbpps'; the subdirectory ./mcbpps/bin/ contains the executable
programs (compiled for 64 bit Linux).

QUICK START: just run the 'run_aaa_example' or the 'run_gtpase_example' scripts
in the examples subdirectory to see how the mcBPPS program works.   

DETAILS: The mcbpps/bin directory contains the following procedures
	 (compiled for Linux x86_64 (64 bit)):

 fa2cma:  Procedure for converting a fasta-formatted multiple sequence alignment 
          into cma-format, which is the format required to create a array of multiple
          sequence alignments--the type of input file required by the mcBPPS procedure. 
          Each cma file follows the convention used by PSI-BLAST, where the first sequence
          serves as a master sequence against which the remaining sequences are aligned.
          Thus each amino acid residue in the first sequence corresponds to a column in the 
          alignment whereas each deletion '-' in the first sequence corresponds 
          to insertions in the alignment. For these reasons, the first sequence is 
          typically a consensus sequence.

 press:   A simple routine for removing from a fasta file extraneous newline characters,
          which fa2cma input files must not contain.

             press command line syntax: press < infile > outfile.

 cma2fa: Procedure for converting a cma-formatted multiple sequence alignment into 
         fasta-format. This is useful for converting the output alignment created by 
         the mcBPPS program into standard fasta format.

 mcBPPS: Details about the main program are as follows:

====================================================================================
===== Multiple category Bayesian Partitioning with Pattern Selection (mcBPPS) =====
====================================================================================

The mcBPPS program: a C++ implementation of a Gibbs sampler for multiple categorization
of functionally divergent residues.

    version 1.0 (June 3, 2010) Copyright 2010, Andrew F. Neuwald

To run demonstrations type the following commands:

cd examples/
./run_aaa_example
./run_gtpase_example

These examples correspond to the P-loop GTPase and AAA+ ATPase examples described in the paper.

Three input files are required by the mcBPPS program:
  1. A file specifying the hyperpartition with optional parameters:     AAAPlus.hpt, PloopGTPase.hpt
  2. A file of concatenated cma-formatted seed multiple alignments:     AAAPlus.sma, PloopGTPase.sma
  3. A file containing the main multiple alignment in cma format:       AAAPlus.mma, PloopGTPase.mma

The command "mcBPPS -" shows the hyperpartition file syntax.

When creating the alignment files the cma2fa and fa2cma programs can be used to convert
fasta formated sequence alignments into *.cma format or vice versa.

The contrast alignments output files generated by the program are in
rich text format (*.rtf), which may be viewed in MicroSoft Word.