Welcome to the ShapeRNA Help Page

This guide will walk you through how to use the ShapeRNA web server for RNA secondary structure prediction and analysis. ShapeRNA provides multiple modes of operation tailored to different input types and analysis goals, including single structure prediction, SHAPE-guided modeling, and structure-based clustering from high-throughput sequencing data.

To begin, please select a module from the top navigation bar or the colored buttons on the homepage. You will then be guided through a step-by-step process—starting with choosing your input format and desired output type.

Single Structure Prediction

This tool allows you to predict a single RNA secondary structure from a plain RNA sequence. It is suitable when no additional experimental data (e.g., SHAPE reactivity) is available and you want to explore a most likely folding based on thermodynamic or learning-based models.

Step 1: Choose the Single Structure Prediction option from the Tool module.

Step 2: Select the prediction model you wish to use (e.g., thermodynamic, deep learning, etc.).

Step 3: Provide your RNA sequence — either by uploading a FASTA file or by typing/pasting the sequence into the input box.

Step 4: Click the Submit button to run the prediction.

Step 5: Once completed, the predicted structure will be displayed and available for download.

Step 6: After the job completes, click the download button to retrieve your predicted structure. (Optional) Use Clear Cache to remove the current results and begin a new task.

Structure Ensemble Prediction

This tool predicts the full ensemble of RNA secondary structures from a plain sequence input. It is suitable when you want to explore the diversity of possible foldings and assess structural uncertainty using base-pairing probabilities or partition function methods, without relying on experimental constraints.

Step 1: Choose the Structure Ensemble option for RNA sequence input.

Step 2: Select the prediction model you wish to use (e.g., thermodynamic, deep learning, etc.).

Step 3: Provide your RNA sequence — either by uploading a FASTA file or by typing/pasting the sequence into the input box.

Step 4: Click the Submit button to run the prediction.

Step 5: Once completed, the predicted structure will be displayed and available for download.

Step 6: After the job completes, click the download button to retrieve your predicted structure.

(Optional) Use Clear Cache to remove the current results and begin a new task.

SHAPE-Guided Single Structure Prediction

This mode allows you to predict a single RNA secondary structure using both sequence and SHAPE reactivity data. SHAPE (Selective 2′-Hydroxyl Acylation analyzed by Primer Extension) data provide experimental constraints that improve the accuracy of structure prediction by guiding the folding process toward biologically relevant conformations.

Step 1: Choose the SHAPE-guided Single Structure option.

Step 2: Select a supported prediction model that accepts SHAPE input.

Step 3: Provide your input by uploading both the RNA sequence (in FASTA format) and the SHAPE reactivity file (plain text), or type/paste the contents into the input box.

Step 4: Drag and drop both files into the upload area or paste their contents directly.

Step 5: Click the Submit button to run the prediction.

Step 6: After the job completes, click the download button to retrieve your predicted structure.

(Optional) Use Clear Cache to remove the current results and begin a new task.

SHAPE-Guided Structure Ensemble Prediction

This mode allows you to generate a structural ensemble of RNA conformations guided by SHAPE reactivity data. By integrating experimental SHAPE constraints into partition function calculations, this method provides a diversity of likely structures with improved biological relevance.

Step 1: Choose the SHAPE-guided Structure Ensemble option.

Step 2: Select a model that supports SHAPE-constrained ensemble prediction.

Step 3: Provide your inputs by uploading both the RNA sequence (FASTA format) and SHAPE reactivity file (plain text), or paste both into the input box.

Step 4: Drag and drop both files into the upload area, or insert data directly.

Step 5: Click the Submit button to begin the ensemble prediction.

Step 6: After processing, click the download button to retrieve your ensemble results.

(Optional) Use Clear Cache to clear previous results before starting a new analysis.

Structure Clustering from FASTQ Data

This mode is designed for analyzing high-throughput sequencing reads in FASTQ format to identify mutation-driven structural heterogeneity. The tool performs structure prediction for variant-containing reads, clusters the resulting structures into distinct structural classes, and outputs representative secondary structures for each cluster.

Step 1: Choose the Structure Clustering from FASTQ option.

Step 2: Select a supported model for mutation-aware structure prediction.

Step 3: Upload your FASTQ file in the designated input section.

Step 4: Drag and drop your file into the upload area or insert your data into the box.

Step 5: Click the Submit button to begin clustering analysis.

Step 6: Download representative structures and cluster summary upon completion.

miRNA Binding Site Analysis

This analysis module identifies potential miRNA binding sites on RNA secondary structures. It takes structured RNA input (in Vienna dot-bracket format) and scans for target motifs based on sequence and accessibility features.

Step 1: Select the miRNA Binding Site Analysis option.

Step 2: Choose relevant parameters such as species or binding rules.

Step 3: Select your input method (e.g., upload file or manual entry).

Step 4: Paste your structured RNA (Vienna format) into the input box.

Step 5: Once the analysis is complete, download the predicted miRNA binding results.

The output of the miRNA Binding Site Analysis consists of a complete table listing all predicted miRNA binding sites on the left, including target sequences, binding positions, and interaction scores. On the right, a ranked summary highlights the top 3 miRNAs with the highest predicted binding confidence. Below, the corresponding binding regions of these top 3 miRNAs are visualized on the RNA secondary structure using an interactive SVG rendered by forna, with the binding sites clearly highlighted for intuitive structural interpretation.

RBP Binding Site Analysis

This module predicts RNA-binding protein (RBP) binding sites based on sequence motifs and RNA structural context.

Step 1: Select the RBP binding analysis mode in the Analysis section.

Step 2: Choose relevant parameters such as species and genome build.

Step 3: Select your input method (upload or manual input).

Step 4: Paste your structured RNA (Vienna format) into the input box.

Step 5: Download your result files use the buttons inside the dashed box.

The output of the RBP Binding Site Analysis module includes a complete table on the upper-left, listing all predicted RBP binding sites along with their matched motifs, binding positions, Z-scores, and P-values. On the upper-right, a ranked summary shows the top 3 RBP binding events, sorted primarily by Z-score, and in cases of ties, prioritized by lower P-values. Below, the RNA secondary structure is visualized using forna, with the binding sites corresponding to the top 3 RBPs highlighted directly on the structure, enabling clear interpretation of their structural context.

m6A Site Prediction

This module identifies potential m6A methylation sites using sequence features and RNA structure.

Step 1: Select the m6A site prediction mode in the Analysis section.

Step 2: Set prediction parameters like species and confidence level.

Step 3: Select your input method.

Step 4: Paste your structured RNA (Vienna format) into the input box.

Step 5: Download your result files use the buttons inside the dashed box.

The output of the m6A Site Prediction module includes a comprehensive table in the upper-left showing all predicted m6A sites, including their positions, motif context, and confidence scores. On the upper-right, a ranked summary displays the top 5 m6A sites sorted by prediction score. In the next row, the left panel presents a secondary structure rendered with forna, where the top 5 predicted sites are highlighted in red on the RNA structure. The right panel shows a structure comparison view, visualizing changes between the original and modified RNA structures under m6A modification at those positions. At the bottom, the updated secondary structure of the RNA is shown with the top 5 sites assumed to be methylated, revealing potential structural shifts caused by m6A incorporation.

Structure Comparison

This tool compares two RNA secondary structures and highlights structural differences.

Step 1: Select the Structure Comparison mode in the Analysis section.

Step 2: Upload two RNA sequences with their dot-bracket structures.

Step 3: Submit to view aligned structures and differences.

Step 4: Download the comparison result if needed.

Step 5: Download your result files use the buttons inside the dashed box.

The output of the Structure Comparison module includes a helix-based visualization that highlights differences between two RNA secondary structures.