Pipeline Parameters

Epi-Flow is controlled by a set of command-line flags. Understanding these parameters is key to tailoring the pipeline to your specific data and research questions.

Required Arguments

These flags are the bare minimum required to start any Epi-Flow run.

Flag	Argument	Description & Guidance
-a	<string>	Assay Type. This is the most important flag, as it tells the pipeline which internal logic to use. Accepted Values: • atac: For ATAC-seq data. Activates Tn5 shift correction and ATAC-specific MACS3 settings. • cutrun: For CUT&RUN data. Activates SEACR peak calling. • cuttag: For CUT&Tag data. Activates Tn5 shifting and SEACR. • chip: For general ChIP-seq data. Uses standard MACS3 settings.
-1	<path>	Path to Read 1 FASTQ file(s). Must be gzipped (.fastq.gz or .fq.gz). For multiple files from the same sample (e.g., from different sequencing lanes), provide them as a single, comma-separated string with no spaces: -1 lane1_R1.fq.gz,lane2_R1.fq.gz
-n	<string>	Sample Name. A unique identifier for your sample. This name will be used as the prefix for all output files. Best Practice: Use a descriptive name without spaces or special characters (underscores are okay). Example: HCT116_H3K27ac_Rep1
-x	<path>	Bowtie2 Index Prefix. The full path to the reference genome index you built in the preparation step. This is the base name of the index files (e.g., /path/to/genomes/hg38).
-o	<path>	Output Directory. The path to a directory where all results will be saved. Note: The pipeline will create this directory if it doesn’t exist.
-g	<integer>	Effective Genome Size. The integer value for the mappable size of your genome. Crucial for accurate peak calling statistics. See the Preparation Guide for a list of common values and how to calculate it for other organisms.

---

Paired-End Data

Flag	Argument	Description & Guidance
-2	<path>	Path to Read 2 FASTQ file(s). This is required for all paired-end assays (atac, cutrun, cuttag, and paired-end chip). The number and order of files must exactly match the files provided with -1: -1 lane1_R1.fq.gz,lane2_R1.fq.gz -2 lane1_R2.fq.gz,lane2_R2.fq.gz

---

Optional Arguments (Tuning & QC)

These flags allow you to customize the analysis, control resource usage, and improve filtering.

Flag	Argument	Default	Description & Guidance
-p	<string>	narrow	Peak Calling Mode. Determines the type of peaks MACS3 will look for. How to choose: • narrow: For sharp, punctate signals like transcription factors (e.g., CTCF), H3K4me3, and H3K27ac. • broad: For diffuse histone marks that cover wide genomic domains (e.g., H3K27me3, H3K36me3, H3K9me3).
-b	<path>	None	Blacklist BED file. The path to a BED file containing artifact-prone genomic regions. Highly Recommended. Using a blacklist file significantly improves the quality of your final peak set by removing false positives.
-t	<integer>	8	Number of Threads. The number of CPU cores to use for parallelizable tasks. Guidance: Set this to the number of cores available on your machine, but it’s often wise to leave one or two free for system processes (e.g., nproc - 2).
-m	<integer>	Auto-detect	Maximum Memory in GB. The total memory available for the pipeline. Epi-Flow automatically detects system RAM, but you can override it on memory-limited systems or HPC nodes to ensure the pipeline requests the correct resources.
--keep-tmp	None	false	Keep Temporary Files. By default, the pipeline cleans up the WORKDIR to save space. This flag prevents the deletion of intermediate files (merged FASTQs, raw BAMs, etc.), which is useful for debugging a failed run.

Putting It All Together: A Real-World Example

This command processes a paired-end ChIP-seq experiment for a broad histone mark, using 16 threads and a blacklist file.

./epi_flow.sh \
    -a chip \
    -n K562_H3K27me3_Rep1 \
    -1 K562_H3K27me3_Rep1_R1.fq.gz \
    -2 K562_H3K27me3_Rep1_R2.fq.gz \
    -x /data/genomes/hg38 \
    -g 2913022398 \
    -b /data/genomes/hg38-blacklist.v2.bed \
    -p broad \
    -t 16 \
    -o ./results/K562_H3K27me3