domains / families / science / secondary-structure-prediction

secondary-structure-prediction

separate

Predicts RNA secondary structure with ViennaRNA, treating the Boltzmann ensemble (partition function, base-pair probabilities, centroid, MEA, stochastic samples) as the object rather than a single MFE fold. Covers consensus folding from alignments (RNAalifold), SHAPE-constrained folding, RNA-RNA interaction (RNAcofold/RNAduplex/RNAup), local and linear-time methods for long RNA, and pseudoknot-aware tools. Use when folding an RNA and choosing between MFE, centroid, MEA, or ensemble sampling; judging whether a single structure is well-defined; folding long RNAs where a global MFE is meaningless; handling suspected pseudoknots; or weighing thermodynamic versus comparative versus deep-learning prediction.

When to use

• folding an RNA and choosing between MFE, centroid, MEA, or ensemble sampling
• judging whether a single structure is well-defined
• folding long RNAs where a global MFE is meaningless
• handling suspected pseudoknots
• weighing thermodynamic versus comparative versus deep-learning prediction
• consensus folding from alignments
• SHAPE-constrained folding
• RNA-RNA interaction

When not to use

• predicting general pseudoknots with nested folders
• predicting non-canonical pairs like sheared G-A or base triples

Skill page

Source