ViennaRNA

ViennaRNA

The ViennaRNA Package consists of a C code library and several stand-alone programs for the prediction and comparison of RNA secondary structures. It is developed and maintained by the Theoretical Biochemistry Group at the University of Vienna.

In case you are using ViennaRNA software for your publications you may want to cite:

Lorenz, Ronny and Bernhart, Stephan H. and Höner zu Siederdissen, Christian and Tafer, Hakim and Flamm, Christoph and Stadler, Peter F. and Hofacker, Ivo L.
ViennaRNA Package 2.0
Algorithms for Molecular Biology, 6:1 26, 2011, doi:10.1186/1748-7188-6-26

RNAlib API Docs

The API docs can be found at https://www.tbi.univie.ac.at/RNA/ViennaRNA/doc/html/index.html

ViennaRNA PyPI Installation

Life at last is simple, and the possibilities are endless.

Just open your favorite shell.

$ pip install ViennaRNA

Sit back, relax and let pip do its thing. When everything is installed, verify like so.

$ python
Python 3.9.5 (default, Jun  4 2022, 12:28:51)
[GCC 7.5.0] :: Anaconda, Inc. on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import ViennaRNA
>>> dir(ViennaRNA)
['RNA', 'RNAlib', 'ViennaRNA', '__author__', '__builtins__', '__cached__', '__doc__', '__file__', '__license__', '__loader__', '__name__', '__package__', '__path__', '__spec__', '__version__', 'main', 'rna', 'rnalib', 'viennarna']

Note The objects RNA, RNAlib, ViennaRNA, rna, rnalib, and viennarna all refer to the same RNA SWIG module that you'd get if you installed the ViennaRNA python interface from source. The statements from ViennaRNA import RNA is then equivalent to from ViennaRNA import rnalib.

Using ViennaRNA from PyPI

The PyPI installer was developed so that computational synthetic biologists could specify ViennaRNA as a requirement for their python applications (via install_requires in their setup.py), without requiring their users to download and install it separately. Because this is the 21st century, and dependency installation should not deter application adoption. Once installed, you can seamlessly import RNA, successfully. Another reason this installer exists is because the conda package for ViennaRNA is broken, and hasn't been fixed in a while. This installer works under conda as well.

Relevant DNA or RNA parameter files will need to be packaged with developed application and supplied to RNA.read_parameter_file(parameter_file_path) for proper usage. You may then want to specify those parameter files as part of your package_data in your setup.py, and make it available to RNA.read_parameter_file(...) via pkg_resource.resource_filename(...) function.

You can find all of the different DNA and RNA parameters accepted by the RNA module in here.

Example: Say, your application is called awesomeRNA and your package is structured like the following.

- awesomeRNA/
    - docs/
    - examples/
    - tests/
    - awesomeRNA/
        - __init__.py
        - awesomeRNA.py
        - utils.py
        - params/
            - dna_mathews2004.par
            - rna_andronescu2007.par
    - setup.py
    - setup.cfg
    - README.md
    - LICENSE

A part of your setup function inside setup.py may then contain the following.

setup(

    # stuff before

    packages=['awesomeRNA', 'awesomeRNA.params'],

    package_dir={
        'awesomeRNA': './awesomeRNA'},

    package_data={
        'awesomeRNA': ['params/*.par']},

    install_requires=[
        'numpy',
        'ViennaRNA']

    # stuff after
)

Then in your application code, you can perhaps do the following.

import RNA
import pkg_resources

def awesomeRNA_main(
    seqeuence,
    temperature=37,
    param_name='dna_matthews2004'):

    # Basic Setup
    if temperature != 37:
        RNA.cvar.temperature = temperature
    RNA.cvar.dangles = 2
    settings = RNA.md()

    # Read Parameters
    parameter_file = pkg_resources.resource_filename(
        'awesomeRNA', 'params/{}.par'.format(
            param_name))
    RNA.read_parameter_file(parameter_file)

    # Calculate MFE and Secondary Structure
    fc_obj = RNA.fold_compound(
        seqeuence,
        settings)
    structure, mfe = fc_obj.mfe()

    # Return Results
    return structure, mfe

Note These examples are for demonstrative purposes only.

Troubleshooting

If you have trouble installing this package inside a conda environment due to libgcc issues, you might want to install mamba (conda install -c conda-forge mamba), or create your environment with mamba to begin with (conda create -n myEnvName -c conda-forge mamba). Then, simply mamba install libgcc and try pip install ViennaRNA again. Additionally, you might require to mamba install libgcc-ng libstdcxx-ng.

License

ViennaRNA Package (c) Theoretical Biochemistry Group, University of Vienna.

ViennaRNA Package has its own custom LICENSE.

ViennaRNA PyPI Installer (c) 2022 Ayaan Hossain.

ViennaRNA PyPI Installer is an open-source software under MIT License.

See LICENSE file for more details.