RNA Secondary Structure Predictor Calculator

Prediction settings

Configure sequence rules, constraints, scoring, and comparison inputs. Server validation checks every submitted option.

RNA sequence or FASTA data

0 nucleotides

Prediction mode

Base-pair rules

Custom pair list

Minimum hairpin loop length

Maximum pairing distance

Use zero for no distance limit.

Maximum accepted length

Temperature

Sodium concentration

M

Magnesium concentration

M

Alternative structure count

Energy-range threshold

Motif search

Allow GU wobble pairs

Allow isolated base pairs

Accept ambiguous IUPAC bases

Treat RNA as circular approximation

Structure constraints

Use x unpaired, | paired, < downstream, > upstream, and . unrestricted.

Required base pairs

Prohibited base pairs

Reference dot-bracket structure

Optional Sequence B

Formula used

The calculator applies a Nussinov dynamic-programming recurrence. It maximizes compatible, noncrossing base-pair scores. Thermodynamic modes remain documented approximations for educational analysis.

N(i,j) = max[N(i+1,j), N(i,j−1), N(i+1,j−1)+δ(i,j), max(N(i,k)+N(k+1,j))]

Here, δ rewards permitted base pairs. Constraints can block or require selected positions. Traceback converts the matrix into dot-bracket notation.

How to use this calculator

Paste an RNA sequence or upload FASTA data.
Select scoring, pairing rules, and loop limits.
Add constraints or a reference structure when needed.
Run prediction and inspect structure warnings.
Review diagrams, alternatives, statistics, and comparison scores.
Export CSV, PDF, CT, BPSEQ, SVG, or PNG.

Example data

RNA sequence	Illustrative notation	Description
GGGAAAUCC	(((...)))	Simple hairpin-like sequence
GCGCUUCGCGC	Structure varies	GC-rich sequence
AAAAAAAA	........	Usually remains unpaired
GGGGCCCC	(((())))	Strong complementary pairing

Prediction limits and scientific use

Computational structures are hypotheses, not experimental confirmations. Proteins, ligands, ions, and kinetics can change folding. Pseudoknots require algorithms beyond standard Nussinov recursion.

This page uses weighted pairing and ensemble heuristics. It does not replace ViennaRNA or laboratory structure mapping. Confirm important conclusions with validated scientific software and experiments.

Frequently asked questions

What does dot-bracket notation show?

Dots mark unpaired nucleotides. Matching brackets mark paired positions. Nested brackets represent pseudoknot-free secondary structures.

Does this calculator predict pseudoknots?

No. The core algorithm produces nested structures only. Reference validation can identify crossing pseudoknot pairs.

What is the minimum loop length?

It limits pairs closing very short hairpins. A value near three is common. Different models may use different restrictions.

What are GU wobble pairs?

Guanine and uracil can form noncanonical interactions. These pairs often appear in functional RNA. Disable them for stricter pairing rules.

Is the reported energy a true MFE?

No. It is a transparent heuristic score. True MFE requires nearest-neighbor thermodynamic parameters.

How does consensus-like mode work?

It requires equal-length sequences. A pair must be compatible in both inputs. It is not a full alignment algorithm.

Why are alternative structures different?

Small deterministic perturbations resolve close scoring choices differently. Alternatives reveal uncertain or variable regions. They are not Boltzmann samples.

Can I compare a known structure?

Yes. Enter a reference dot-bracket string with matching length. The calculator reports sensitivity, PPV, F1, and distance.

Which export formats are supported?

Results export as CSV, PDF, TXT, CT, BPSEQ, SVG, and PNG. Printing creates a clean report. Copying captures the main prediction summary.