RNA structure · Software
ViennaRNA (RNAfold)
Predicts and compares RNA secondary structures and their thermodynamics for computational and molecular biologists.
What ViennaRNA (RNAfold) is
The ViennaRNA Package is a C code library and suite of stand-alone command-line programs designed for predicting and analysing RNA secondary structures. Originating from researchers including Ivo Hofacker and Ronny Lorenz at the University of Vienna's Theoretical Biochemistry Group, the software addresses the computational challenge of determining how RNA sequences fold into stable configurations and how these structures interact.
The software provides utilities for calculating minimum free energy conformations, computing equilibrium base-pairing probabilities via partition functions, and determining suboptimal structures. It also supports consensus structure prediction across multiple sequence alignments, RNA-RNA hybridization modeling, and inverse folding for sequence design.
For molecular researchers, understanding secondary structure is essential for studying RNA function, regulatory mechanisms, and molecular interactions. ViennaRNA handles long sequences, circular RNAs, G-quadruplexes, modified bases, and experimental probing constraints from methods such as SHAPE.
How it works
ViennaRNA uses dynamic programming algorithms to compute thermodynamic properties. Secondary structure prediction operates primarily through three methods: the minimum free energy algorithm of Zuker and Stiegler to find a single optimal structure, McCaskill's partition function algorithm to calculate base-pairing probabilities across the thermodynamic ensemble, and the suboptimal folding algorithm of Wuchty et al. to generate alternative configurations within a defined energy range. Global structure prediction uses an M2 dynamic programming matrix decomposition that accounts for at least two branches within a multiloop.
Calculations rely on nearest-neighbour thermodynamic energy parameters, defaulting to Turner 2004 parameters alongside options for Turner 1999, trained sets, and DNA parameters. The package also incorporates energy models for modified bases, physics-based salt corrections for mono-valent ion concentrations, and soft or hard constraint frameworks that incorporate experimental chemical probing data.
Key facts
- Primary language
- C
- Language interfaces
- Perl 5 and Python 3
- Input formats
- Standard input and FASTA format
- Output visualization formats
- PostScript, SVG, and GML
- Maximum sequence length
- Up to 32,700 nucleotides
- Default energy parameters
- Turner 2004
- Current version
- 2.7.2
What you can use it for
- Minimum free energy secondary structure prediction
- Predict the most thermodynamically stable secondary structure and free energy for a given RNA sequence using RNAfold.
- Thermodynamic ensemble and base-pairing analysis
- Calculate partition functions and equilibrium base-pairing probabilities across ensembles to evaluate structural stability and melting behaviour.
- Consensus folding of aligned sequences
- Determine conserved consensus secondary structures from multiple sequence alignments using RNAalifold or RNALalifold.
- RNA-RNA interaction and hybridization modeling
- Model binding interactions and complex formation across two or multiple interacting RNA strands using RNAcofold, RNAduplex, or RNAmultifold.
- RNA sequence design
- Design RNA sequences that fold into a target secondary structure using inverse folding algorithms in RNAinverse.
- Experimental constraint integration
- Incorporate chemical probing constraints, such as SHAPE reactivity profiles, into secondary structure predictions as hard or soft constraints.
Run it here
ViennaRNA (RNAfold) is preinstalled on the machine behind every research conversation. Ask the agent in plain language, or give it the exact command — it runs it, reads the output and explains it.
Strengths and limits
Good at
- Provides multiple established dynamic programming algorithms for optimal, suboptimal, and ensemble partition function calculations.
- Includes extensive language bindings for both Python 3 and Perl 5 to access the underlying RNAlib C library.
- Supports specialized structural features including G-quadruplexes, circular RNAs, modified bases, and mono-valent salt corrections.
- Command-line utilities follow UNIX conventions by reading standard input and writing to standard output for pipeline integration.
Watch out for
- Predictions for sequences approaching the upper limit of 32,700 nucleotides require very large amounts of system memory.
- Energy parameters for modified bases remain incomplete and are largely restricted to base pair stacking contributions.
- Legacy utilities for analysing sequence and distance data using split decomposition and cluster methods are unmaintained and omitted from default builds.
Questions
Can I run ViennaRNA (RNAfold) online without installing it?
ViennaRNA (RNAfold) is preinstalled on the machine behind every research conversation. Ask the agent in plain language, or give it the exact command — it runs it, reads the output and explains it.
What licence is ViennaRNA (RNAfold) under?
ViennaRNA (RNAfold) is published under its authors' own licence; the project page at https://www.tbi.univie.ac.at/RNA/ states it.
Can ViennaRNA predict structures for sequences containing modified bases?
Yes, it supports modified bases such as inosine, pseudouridine, m1Ψ, m5C, m6A, 7DA, purine, and dihydrouridine, though parameters are mostly limited to base pair stacking.
Does the package support DNA folding?
Yes, secondary structure predictions can be performed for DNA by selecting included DNA parameter sets such as Mathews 1999 or Mathews 2004.
How are experimental probing data incorporated into predictions?
Tools such as RNAfold, RNAalifold, and sliding-window tools support generic hard- and soft-constraint frameworks that directly integrate chemical probing data such as SHAPE reactivities.
Can ViennaRNA model interactions among more than two RNA molecules?
Yes, the RNAmultifold program computes secondary structures, partition functions, and equilibrium concentrations for complexes formed by multiple interacting RNA strands.
Which programming interfaces are provided for development?
The package provides the C library RNAlib along with official bindings for Python 3 and Perl 5.
How to cite it
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
Written from ViennaRNA (RNAfold)'s own README and project page. Statements citing a figure that does not appear there are removed before publication. Licence and running details come from our registry.