This repository contains Nextflow-based analysis tools for Enzymatic Methylation Sequencing (EM-seq) and Enzymatic 5hmC-seq (E5hmC-seq) data processing.

Complete EM-seq processing pipeline that accepts UBAM inputs:

• Adapter trimming and read alignment with (fastp, bwa-meth)
• Duplicate marking (Picard)
• Methylation calling (MethylDackel)
• Quality control metrics and statistics (Picard, Samtools, FastQC, MultiQC)
• Optional BED file intersection for targeted analysis (bedtools)

If your files are in fastq format, you will need to convert them to uBams prior to running the main pipeline, e.g.:

nextflow run fastq_to_ubam.nf \
  --input_glob "tests/fixtures/fastq/emseq-test*{.ds.1,.ds.2}.fastq.gz" \
  --read_format 'paired-end'

1. Install miniforge and bioconda (see Requirements)
2. Install Nextflow (e.g. conda install nextflow, or see Nextflow installation guide)
3. Clone this repository (git clone https://github.com/nebiolabs/EM-seq.git). Modify nextflow.config as needed for your environment, e.g. if running locally, change executor block to 'local' and set, e.g. --max_cpus 10 --max_memory 30.GB.
4. Download or prepare a genome reference FASTA file (see Reference Genomes)
5. Create a bwameth index for the fasta and add it to your references in conf/references.config
6. Run the pipeline with appropriate parameters (see Basic Usage)
7. Examine results in the EM-seq_output directory
   • EM-seq-Alignment-Summary-<FLOWCELL_ID>_multiqc_report.html in em-seq_output for overall QC summary
   • Mbias files em-seq_output/methylDackelExtracts/mbias (to identify sample-dependent positional biases)
   • Methylation output files in em-seq_output/methylDackelExtracts (suitable for analysis with methylKit)
   • Aligned reads in em-seq_output/markduped_bams (methylation coloring is recommended for visualization in IGV)

nextflow run main.nf \
  --genome 'test' \
  --ubam_dir './' \
  --email your.email@example.com \
  --flowcell FLOWCELL_ID

ubam_dir should be the folder where your ubam files are.

Modify the conf/references.config file to specify your genome files

• genome_fa path to your genome fasta file
• genome_fai path to your genome fasta fai file
• bwameth_index path to your genome fasta file where bwameth indices exist
• target_bed BED file for targeted analysis, Optional

• --tmp_dir - Temporary directory (default: /tmp)
• --workflow - Workflow identifier (default: EM-seq)
• --enable_neb_agg - Enable NEB aggregation reporting (default: False)

Pre-built reference genomes with methylation spike-in controls:

• T2T CHM13: https://neb-em-seq-sra.s3.amazonaws.com/T2T_chm13v2.0%2Bbs_controls.fa
• GRCh38: https://neb-em-seq-sra.s3.amazonaws.com/grch38_core%2Bbs_controls.fa
• Create your own reference by appending the control sequences to your preferred genome fasta (e.g. cat genome.fa methylation_controls.fa > genome+methylation_controls.fa)

  Sequence	Methylation State	Purpose
  lambda	All Cs are unmodified (included in kits)	confirmation that APOBEC enzyme is working optimially (unprotected C->T)
  pUC19c	All CpG sites contain 5mC (included EM-seq kits)	confirmation that TET2 + T4-BGT step is protecting 5mC (5mC -> 5ghmC/5caC
  T4	All Cs are 5hmC (included in the 5-hmC Seq kits)	confirmation that T4-BGT is protecting 5hmC optimally (5hmC -> 5ghmC)

• Nextflow
• Miniforge, Micromamba, or Conda for dependency management
• Bioconda channel configured
• Sufficient computational resources (memory scales with input size)

These in the "legacy" folder are retained for reference and reproducibility but are not actively maintained and are not compatible with the latest Nextflow versions. Use NXF_VER=22.10.4 nextflow run ... to reproduce the results in the EM-seq paper.

• em-seq.nf - Original alignment and methylation calling workflow
• bins.nf - TSS-centered binned coverage analysis
• cov_vs_meth.nf - Coverage vs methylation analysis for genomic features

Analysis methods in this repository were used in the following publication:

Vaisvila R, Ponnaluri VKC, Sun Z, et al. Enzymatic methyl sequencing detects DNA methylation at single-base resolution from picograms of DNA. Genome Res. 2021;31(7):1280-1289. doi:10.1101/gr.266551.120

You may also be interested in the nf-core methylseq project

• git tag -f current_production
• git push -f origin current_production

• development workflow will run from master branch

• Tests are run using nf-test and are integrated into github actions
• install nf-test from bioconda using conda/mamba
• To run all tests:

nf-test test

• When new tests are added or results change, to update the results snapshot:

nf-test test --updateSnapshot