README

Last updated Nov 8, 2018.  Includes "consolidation" work done by Georgia Tech REU student Harrison Wilco (Summer 2018).  

Not actively maintained! 

RNAStructProfilingConsolidation

Compiler dependencies: To compile and install RNAStructProfilingConsolidation, you will need
--a C/C++ compiler
--autotools
--GMP (version 6.0.0 from gmplib.org)
Parallelization requirements: For parallelization, the compiler needs to implement the OpenMP API.
(GNU gcc 4.7, for example, implements the OpenMP API. See http://openmp.org/wp/openmp-compilers for details.)
Graphing requirements: Graphviz must be installed

To download source code, click on "zip" link on the RNAStructProfilingConsolidation github
page (https://github.gatech.edu/hwilco6/RNAStructProfilingConsolidation).

To compile on a Mac OSX machine, make sure your dependencies are satisfied.
(The easiest way to satisfy dependencies is to install Xcode command line tools,
the gcc compiler 4.7 or higher using MacPorts, and possibly GMP 6.0.0 if needed.)
Rename
libgtfold_osx.a -> libgtfold.a
and follow the rest of directions.

To compile on a Unix/Linux machine, make sure the following files are in the
same directory:
--array_list.c
--array_list.h
--consolidated_graph.c
--consolidated_graph.h
--graph.c
--graph.h
--hashtbl.c
--hashtbl.h
--helix_class.c
--helix_class.h
--libgtfold.a
--list.h
--main.c
--memoryDFS.c
--memoryDFS.h
--Options.c
--Options.h
--profile.c
--profile.h
--Profnode.c
--Profnode.h
--Set.c
--Set.h
--stem.c
--stem.h
--Makefile
--include folder with header files
--data folder with GTparams

Open a terminal window, navigate to the directory that contains
the above files, and type 'make' to compile the program.

To run, you will need to input a sequence file, and optionally a
structure file either in sfold (using -sfold option) or in gtboltzmann 
(using -e option) format.

Usage (from within that same directory): 
./RNAprofile [options] [seq file] 

To generate a jpeg from the graphviz dot file made using -o [file name]
dot -T jpeg -o [graph file].jpeg [file name]

Common options:
-v = verbose output that gives all helix classes and profiles with frequencies
-gc = set maximum graph size in number of vertices. -1 for no limit
-h [percent] = filter out all helix classes that occur in less than [h] percent
 of the input sample structures
-p [percent] = filter out all profiles that occur less than [p] percent of the
 input sample structures
-g = disable graph making component of program (if only interested in helix and
 profile stats)
-o [file name] = produces the output graphviz dot files called [file name].dot
 and [file name]_consolidated.dot; 'profile.dot' used by default
-e [file] gtboltzmann style structure file as input
-sfold [file] Sfold style structure file as input

Consolidation options:
-sc = run without consolidation
-sg = disable graph making component of consolidation (if only interested in
 stem and stem profile stats)
-snh = max number of helices to consider for initial stem creation
-sh [percent] = filter out all stems that occur in less than [sh] percent of the
 input sample structures
-sp  = [percent] = filter out all stem profiles that occur less than [sp] percent
 of the input sample structures

For a more extensive list of options, run ./RNAprofile --help