Predict RNA 2D Structure

A Nextflow pipeline for transcriptome-scale RNA secondary structure prediction and annotation. The pipeline accepts transcriptome FASTA input and optional SHAPE and exon annotation files, then runs RNA structure prediction in a reproducible HPC/container environment.

Features

• Transcriptome-wide RNA secondary structure prediction
• FASTA-based transcript input
• Optional SHAPE reactivity integration
• Optional exon annotation integration
• Parallel processing through transcript splitting
• Reproducible execution with Nextflow and Singularity/Apptainer
• Resume support for interrupted runs

Repository structure

.
├── bin/                         # Helper scripts
├── conf/                        # Nextflow configuration files
├── lib/                         # Pipeline libraries
├── modules/local/               # Local Nextflow modules
├── test/                        # Test input files
├── workflows/                   # Workflow definitions
├── main.nf                      # Main Nextflow entry point
├── nextflow.config              # Default configuration
├── run_rna_structure_prediction.sh
└── test.sh

Requirements

• Linux or HPC environment
• Nextflow
• Singularity or Apptainer
• RNA structure tools configured in nextflow.config, such as RNAfold/RNAplot

On many HPC systems, Singularity can be loaded with:

module load singularity

Input files

Input	Description
--transcriptome_fa	Transcriptome FASTA file. Each transcript should have a unique FASTA header.
--exon_data	Exon annotation data in BED-like format.
--shape_data	Optional SHAPE reactivity data in BED-like format.
--split_num	Number of chunks used to split transcriptome sequences for parallel processing.

Quick start

module load singularity

BASEPATH=/path/to/predict_rna_structure/test
WORKDIR=$BASEPATH/work_test
OUTDIR=$BASEPATH/results_test
TRANSCRIPTOME_FASTA=$BASEPATH/rawdata/test_m.fa
shape_data=$BASEPATH/rawdata/test_shape.bed
exon_data=$BASEPATH/rawdata/test_exon.bed

current_datetime=$(date "+%Y%m%d%H%M%S")
LOGPATH=$BASEPATH/log
mkdir -p $WORKDIR $OUTDIR $LOGPATH

nextflow -log $LOGPATH/${current_datetime}.log \
    run main.nf \
    --outdir $OUTDIR \
    -work-dir $WORKDIR \
    --transcriptome_fa $TRANSCRIPTOME_FASTA \
    --split_num 100 \
    --shape_data $shape_data \
    --exon_data $exon_data \
    -resume

Output

Results are written to the directory specified by:

--outdir

The output directory contains final RNA structure prediction results and intermediate files generated by the enabled workflow modules.

Resume or clean a run

Resume an interrupted run:

nextflow run main.nf -resume

Clean Nextflow cache:

nextflow clean -f

Configuration

Default parameters and execution profiles are defined in nextflow.config. Common profiles include:

• singularity
• apptainer
• docker
• conda
• mamba
• debug

Tool paths such as RNAfold, RNAplot, b2mt.pl, and bpRNA.pl should be configured in nextflow.config for the target computing environment.

Citation

If you use this pipeline, please cite the corresponding RBrowser manuscript and application.

Contact

Lei Zheng
GitHub: typekey
Email: type.zheng@gmail.com