Nanopore DNA modification analysis pipeline

---

Overview

ShadowBase is a Nextflow DSL2 pipeline for analyzing DNA modifications from Nanopore sequencing data. It supports both single-sample and multi-sample analysis, where each sample can use a different reference sequence.

Pipeline steps

1. mod_mapping — Remaps MM/ML modification tags from BAM to reference via minimap2
2. modkit pileup — Generates pileup BED with modification frequencies
3. modkit find-motifs — Identifies enriched sequence motifs
4. modkit call-mods — Produces modification-called BAM
5. modkit extract — Extracts modification calls to TSV
6. MicrobeMod (optional) — Motif calling and RM gene annotation
7. MethylomeMiner (optional) — Methylome analysis with GFF annotations

---

Prerequisites

• Java 8+ (OpenJDK / default-jre)
• Nextflow 20.10.0+
• Container engine — either Docker or Singularity

# Install Java
sudo apt install -y default-jre

# Install Nextflow
curl -s https://get.nextflow.io | bash
sudo mv nextflow /bin && sudo chmod 770 /bin/nextflow

---

Quick start

Single sample

nextflow run main.nf \
  --fasta_ref ref.fasta \
  --bam sample.bam \
  --output results \
  -profile local,docker

Multiple samples (same or different references)

Create a CSV:

SampleA,/path/to/sampleA.bam,/path/to/refA.fasta
SampleB,/path/to/sampleB.bam,/path/to/refB.fasta

Then run:

nextflow run main.nf \
  --samples samples.csv \
  --output results \
  -profile local,docker

With per-sample annotations (optional 4th column)

SampleA,/path/to/sampleA.bam,/path/to/refA.fasta,/path/to/annotationA.gff
SampleB,/path/to/sampleB.bam,/path/to/refB.fasta

Annotations enable the MethylomeMiner step.

Test data

nextflow run main.nf \
  --fasta_ref test_data/fastas/A_conjugant_0.5h_polished.fasta \
  --bam test_data/subsampled_bam/A_conjugant_0.5h_10percent.bam \
  --output test_output \
  -profile local,docker

---

Parameters

Param	Default	Description
--fasta_ref	—	Reference FASTA (single sample mode)
--bam	—	BAM with MM/ML modification tags (single sample mode)
--samples	—	CSV file (name,bam,fasta[,annotation_gff])
--output	results	Output directory
--cores	max_cores/4	CPUs per process
--max_cores	all CPUs	Machine-wide CPU limit
--gff_annotation	—	Global GFF annotation for MethylomeMiner
--no_microbemod	false	Skip MicrobeMod (motif calling + RM genes)
--filter_threshold_modkit	—	modkit pileup filter threshold

---

Profiles

Always use a two-part profile: -profile <Executor>,<Engine>

Executor	Engine
local	docker
slurm	singularity
lsf
ukj_cloud

nextflow run main.nf ... -profile local,docker
nextflow run main.nf ... -profile slurm,singularity

---

Output structure

results/
├── Pipeline-runinfo/
│   ├── execution_timeline.html
│   └── execution_report.html
├── <sample_name>/
│   ├── 1.Modmapped_bam/
│   │   ├── <sample>.modmapped.bam
│   │   └── <sample>.modmapped.bam.bai
│   ├── 2.Microbemod_motifs/        (if --no_microbemod is not set)
│   │   ├── <sample>_motifs.tsv
│   │   ├── <sample>_methylated_sites.tsv
│   │   └── RM_genes/
│   ├── 3.Modkit/
│   │   ├── <sample>_pileup.bed
│   │   ├── find_motifs/
│   │   │   ├── <sample>_motifs.tsv
│   │   │   └── <sample>_modkit_find_motifs_log.txt
│   │   ├── call-mods-bam/
│   │   │   └── <sample>_call-mods.bam
│   │   └── extract/
│   │       └── <sample>_extract.tsv
│   └── 4.MethylomeMiner/           (if GFF annotation provided)
│       └── <sample>_MethylomeMiner_output/

---

Acknowledgments

Developed by Mike Marquet — @mult1fractal

• modkit — Nanopore modified base analysis toolkit
• MicrobeMod — Microbial methylation analysis
• Nextflow — Workflow orchestration

License

GPL-3.0