methSV is a robust and scalable Nextflow-based framework for extracting and analyzing DNA methylation signals within structural variation (SV) regions using long-read sequencing data.
- Processes, filters, and standardizes SV VCF files for downstream methylation analysis
- Extracts CpG methylation signals from heterozygous deletions (DELs) recorded in the VCF
- Retrieves insertion (INS)-associated CpG methylation signals directly from the BAM file
methSV/
├── nextflow/ # Main Nextflow pipeline and modules
├── demo/ # Demo dataset for quick testing
├── data/ # Published results used in our study
│ ├── exp_statistic/ # Extracted methylation experimental results
│ ├── exp_pDMR/ # pDMR (population DMR) experimental results
│ ├── exp_hDMR/ # hDMR (heterozygous DMR) experimental results
│ └── exp_enrichment/ # Functional enrichment analysis results
├── work/ # Nextflow run result instance
└── shell/ # Shell scripts for preprocessing and analysis
git clone https://github.com/YLeeHIT/methSV.git
cd methSVcurl -s https://get.nextflow.io | bash
mv nextflow ~/bin/Or use ./nextflow directly in the repo if not moved to your PATH.
A minimal test dataset is provided in the demo/ directory. You can test the pipeline as follows:
cd nextflow
nextflow run methSV.nf \
--input_vcf ../demo/sam1.cuteSV_force_calling.genotype.vcf.gz \
--sample_id sam1 \
--threads 4 \
--methylation_bed ../demo/genome.pos \
--input_bam ../demo/sam1_chr22_head-5000.bam
To run methSV on your own dataset, use the following command:
nextflow run methSV.nf \
--input_vcf <your_vcf.gz>
--sample_id <sample_ID> \
--threads <number of threads> \
--methylation_bed <*bed> \
--genome_file <genome_features.pos> \
--input_bam <your_bam> Required parameters:
| Parameter | Description |
|---|---|
--input_vcf |
Structural variant file (VCF format), e.g., from cuteSV |
--sample_id |
Unique identifier for the input sample |
--threads |
Number of threads to use |
--methylation_bed |
BED file containing CpG methylation calls |
--genome_file |
Genome feature position file (e.g., for computing distance) |
--input_bam |
BAM file aligned to reference genome using ONT reads |
Before running the core methSV pipeline, users must preprocess the raw sequencing data to extract structural variants (SVs) and base-resolution DNA methylation profiles.
We have also provide our own preprocessing scripts:
| Script Name | Description | Example Command |
|---|---|---|
download_giab_pod5.sh |
Download GIAB HG002–HG007 pod5 files from ONT S3 bucket | bash download_giab_pod5.sh |
dorado_basecaller_align_methcall.sh |
Perform basecalling, alignment, and methylation calling | bash dorado_basecaller_align_methcall.sh <sample_id> |
samtools_sort_index_bam.sh |
Sort and index a BAM file | bash samtools_sort_index_bam.sh <input.bam> <threads> |
samtools_filter_mapq.sh |
Filter reads from BAM file | bash samtools_filter_mapq.sh <input.bam> <output.bam> [threads] |
modkit_pileup_bam.sh |
Extract base-resolution methylation from BAM | bash modkit_pileup_bam.sh <input_bam> <output_bed> <reference_fasta> |
sniffles_callSV_bam.sh |
Call SVs from a BAM file | bash sniffles_callSV_bam.sh <input.bam> <output.vcf> <threads> <reference.fa> |
clair_callSNV_bam.sh |
Call SNVs and filter | bash clair_callSNV_bam.sh <input_bam> <sample_id> <threads> <ref_genome> <model_path> <clair3_script> <output_dir> |
metilene_dmr.sh |
Identify DMRs between two groups using metilene | bash metilene_dmr.sh <group1> <group2> <root_dir> <threads> |
- Single-threaded Data Processing
- Structural Variant Preprocessing
- Methylation Extraction for Deletions
- Methylation Extraction for Insertions
If you use methSV in your work, please cite:
Li Y. et al. "methSV: A Long-Read-Based Framework for Profiling DNA Methylation in Structural Variation" (in preparation, 2025)
For any questions, please contact email