ViennaRNA Package

A C code library and several stand-alone programs for the prediction and comparison of RNA secondary structures.

Amongst other things, our implementations allow you to:

• predict minimum free energy secondary structures
• calculate the partition function for the ensemble of structures
• compute various equilibrium probabilities
• calculate suboptimal structures in a given energy range
• compute local structures in long sequences
• predict consensus secondary structures from a multiple sequence alignment
• predict melting curves
• search for sequences folding into a given structure
• compare two secondary structures
• predict interactions between multiple RNA molecules

The package includes Perl 5 and Python modules that give access to almost all functions of the C library from within the respective scripting languages.

There is also a set of programs for analyzing sequence and distance data using split decomposition, statistical geometry, and cluster methods. They are not maintained any more and not built by default.

The code very rarely uses static arrays, and all programs should work for sequences up to a length of 32,700 (if you have huge amounts of memory that is).

See the NEWS and CHANGELOG.md files for changes between versions.

Availability

The most recent source code should always be available through the official ViennaRNA website and through github.

Documentation

Executable programs shipped with the ViennaRNA Package are documented by corresponding man pages, use e.g.:

man RNAfold

in a UNIX terminal to obtain the documentation for the RNAfold program. HTML translations of all man pages can be found at our official homepage.

We maintain a reference manual describing the RNAlib API that is automatically generated with doxygen. The HTML version of this reference manual is available at our official website and at ReadTheDocs.

Installation

For best portability the ViennaRNA package uses the GNU autoconf and automake tools. The instructions below are for installing the ViennaRNA package from source.

See the file INSTALL for a more detailed description of the build and installation process.

Quick Start

Usually you'll simply unpack the distribution tarball, configure and make:

tar -zxvf ViennaRNA-2.7.1.tar.gz
cd ViennaRNA-2.7.1
./configure
make
sudo make install

User-dir Installation

If you do not have root privileges on your computer, you might want to install the ViennaRNA Package to a location where you actually have write access to. Use the --prefix option to set the installation prefix like so:

./configure --prefix=/home/username/ViennaRNA
make install

This will install everything into a new directory ViennaRNA directly into the home directory of user username.

Note, that the actual install destination paths are listed at the end of the ./configure output.

Install from git repository

If you attempt to build and install from our git repository, you need to perform some additional steps before actually running the ./configure script:

1. Unpack the libsvm archive to allow for SVM Z-score regression with the program RNALfold:

cd src
tar -xzf libsvm-3.35.tar.gz
cd ..

2. Unpack the dlib archive to allow for concentration dependency computations with the program RNAmultifold:

cd src
tar -xjf dlib-20.0.tar.bz2
cd ..

3. Install the autotools toolchain and the additional maintainer tools gengetopt, help2man,flex,xxd, and swig if necessary. For instance, in Debian based distributions, the following packages need to be installed:

   • build-essential (basic build tools, such as compiler, linker, etc.)
   • autoconf, automake, libtool, pkg-config (autotools toolchain)
   • gengetopt (to generate command line parameter parsers)
   • help2man (to generate the man pages)
   • bison and `flex`` (to generate sources for RNAforester)
   • vim-common (for the xxd program)
   • swig (to generate the scripting language interfaces)
   • liblapacke (for RNAxplorer)
   • liblapack (for RNAxplorer)
   • A fortran compiler, e.g. gfortran (for RNAxplorer)

4. Finally, run the autoconf/automake toolchain:

autoreconf -i

After that, you can compile and install the ViennaRNA Package as if obtained from the distribution tarball.

Binary packages

Binary packages for several Linux-based platforms, Microsoft Windows, and Mac OS X are available at our official website.

Bioconda

Installation is also possible through bioconda. After successfully setting up the bioconda channels

conda config --add channels defaults
conda config --add channels bioconda
conda config --add channels conda-forge
conda config --set channel_priority strict

you can install the viennarna bioconda package through

conda install viennarna

Python interface only

The Python 3 interface for the ViennaRNA Package library is available at PyPI and can be installed independently using Python's pip:

python -m pip install viennarna

Building a Python 3 sdist or wheel package

Our source tree allows for building/installing the Python 3 interface separately. For that, we provide the necessary packaging files pyproject.toml, setup.cfg, setup.py and MANIFEST.in.

These files are created by our autoconf toolchain after a run of ./configure. Particular default compile-time features may be (de-)activated by setting the corresponding boolean flags in setup.cfg. See below for additional steps when building the Python interface from a clean git clone.

Running

python -m build

will then create a source distribution (sdist) and a binary package (wheel) in the dist/ directory. These files can be easily installed via Python's pip.

Howto prepare the Python 3 sdist/wheel build from git repository

If you are about to create the Python interface from a fresh clone of our git repository, you require additional steps after running ./configure as described above. In particular, some autogenerated static files that are compiled into RNAlib must be generated. To do so, run

cd src/ViennaRNA/static
make
cd ../../..

Additionally, if building the reference manual is not explicitly turned off, the Python interface requires docstrings to be generated. They are taken from the doxygen xml output which can be created by

cd doc
make refman-html
cd ..

Finally, the swig wrapper must be build using

cd interfaces/Python
make RNA/RNA.py
cd ../..

After these steps, the Python sdist and wheel packages can be build as usual.

References

If you use our software package, you may want to cite the follwing publications:

• R. Lorenz et al. (2011), "ViennaRNA Package 2.0", Algorithms for Molecular Biology, 6:26

• I.L. Hofacker (1994), "Fast folding and comparison of RNA secondary structures", Monatshefte fuer Chemie, Volume 125, Issue 2, pp 167-188

Note, that the individual executable programs state their own list of references in the corresponding man-pages.

License

Please read the copyright notice in the file COPYING!

If you're a commercial user and find these programs useful, please consider supporting further developments with a donation.

Contact

We need your feedback! Send your comments, suggestions, and questions to rna@tbi.univie.ac.at

Ivo Hofacker, Spring 2006