Execute the ENCODE ATAC-seq processing pipeline from raw FASTQ files through Tn5 offset\ncorrection, peak calling, IDR analysis, and signal track generation. This skill provides\na Nextflow DSL2 implementation following ENCODE uniform analysis standards.
\nTSS enrichment scoring is a manual post-processing step; the workflow does not compute\nit (see \"Manual QC steps\" below and references/05-qc-metrics.md).
ATAC-seq (Assay for Transposase-Accessible Chromatin using sequencing) uses the Tn5\ntransposase to probe open chromatin regions. This pipeline processes ATAC-seq data\nthrough quality control, alignment with Bowtie2, mitochondrial read removal, duplicate\nremoval, Tn5 insertion site correction (+4/-5 bp offset), blacklist filtering,\nnucleosome-free fragment selection, MACS2 peak calling, FRiP calculation, and an IDR\ncomparison for every pair of replicates.
\nKey differences from ChIP-seq: Bowtie2 aligner (optimized for short fragments), Tn5\ntransposase shift correction, mitochondrial read filtering (chrM can be 30-80% of reads),\nand no input control.
\nThe workflow is paired-end only. Passing --single_end stops the run with an error,\nbecause Tn5 shifting, nucleosome-free selection and BAMPE peak calling all depend on\nfragment length.
| Reference | \nJournal | \nYear | \nDOI | \nRelevance | \n
|---|---|---|---|---|
| Buenrostro et al. \"Transposition of native chromatin (ATAC-seq)\" | \nNature Methods | \n2013 | \n10.1038/nmeth.2688 | \nOriginal ATAC-seq method (~5,000 citations) | \n
| Corces et al. \"An improved ATAC-seq protocol\" | \nNature Methods | \n2017 | \n10.1038/nmeth.4396 | \nOmni-ATAC improvements (~2,500 citations) | \n
| ENCODE Project Consortium \"Expanded encyclopaedias\" | \nNature | \n2020 | \n10.1038/s41586-020-2493-4 | \nENCODE Phase 3 standards | \n
| Amemiya et al. \"ENCODE Blacklist\" | \nScientific Reports | \n2019 | \n10.1038/s41598-019-45839-z | \nArtifact regions (~1,372 citations) | \n
| Langmead & Salzberg \"Fast gapped-read alignment with Bowtie 2\" | \nNature Methods | \n2012 | \n10.1038/nmeth.1923 | \nAligner (~30,000 citations) | \n
| Yan et al. \"From reads to insight: ATAC-seq analysis\" | \nGenome Biology | \n2020 | \n10.1186/s13059-020-1929-3 | \nAnalysis best practices | \n
FASTQ ──> FastQC / Trim Galore ──> Bowtie2 ──> Mito Removal ──> Picard MarkDuplicates\n │ (chrM dropped) (duplicates REMOVED)\n │ │\n │ ┌───────────────────────────────────────────────────────────┘\n │ v\n │ Tn5 Shift (alignmentSieve --ATACshift) ──> Blacklist Filter ──> Size Selection\n │ │ │\n │ │ ┌───────────┴────────┐\n │ v v v\n │ Signal Track NFR (<150 bp) Mono-nucleosome\n │ (all frags) │ (150-300 bp)\n │ v\n │ MACS2 Peak Calling ──> IDR (every pair)\n │ │\n │ v\n │ FRiP (NFR peaks vs final BAM)\n v\n QC reports ────────────────────────────────────────────────────────────────> MultiQC\nThe Tn5 shift runs after duplicate removal, and peaks are called on the\nnucleosome-free BAM only. The signal track is built from all fragments in the\nblacklist-filtered BAM, not from the NFR BAM.
\n| Stage | \nTool | \nInput | \nOutput | \nReference | \n
|---|---|---|---|---|
| 1. QC & Trimming | \nFastQC, Trim Galore | \nRaw FASTQ | \nTrimmed FASTQ, FastQC reports | \nreferences/01-qc-trimming.md | \n
| 2. Alignment | \nBowtie2, samtools | \nTrimmed FASTQ | \nSorted BAM, flagstat, bowtie2 log | \nreferences/02-alignment.md | \n
| 3. Filtering & Tn5 shift | \nsamtools, Picard, deeptools alignmentSieve, bedtools | \nSorted BAM | \nShifted, filtered, size-selected BAMs | \nreferences/03-tn5-filtering.md | \n
| 4. Peak Calling & IDR | \nMACS2, IDR | \nNFR BAM | \nnarrowPeak, one <sampleA>_vs_<sampleB>.idr_peaks.txt per replicate pair | \nreferences/04-peak-calling.md | \n
| 5. Signal, FRiP & QC report | \ndeeptools bamCoverage, bedtools, samtools, MultiQC | \nFiltered BAM, NFR peaks, QC logs | \nbigWig, <sample>.frip_mqc.tsv, multiqc_report.html | \nreferences/05-qc-metrics.md | \n
--reads): paired-end reads, gzipped. A Nextflow file-pair glob,\ne.g. 'fastq/*_R{1,2}.fq.gz'.--bowtie2_index): Bowtie2 is invoked as\nbowtie2 ... -x <dir>/<genome>, so the directory must hold index files named after the\ngenome:GRCh38_bowtie2_index/\n GRCh38.1.bt2 GRCh38.2.bt2 GRCh38.3.bt2 GRCh38.4.bt2\n GRCh38.rev.1.bt2 GRCh38.rev.2.bt2\nBuild it once with bowtie2-build GRCh38.fa GRCh38_bowtie2_index/GRCh38. If the flag is\nomitted, the workflow looks for ./<genome>_bowtie2_index in the launch directory. The\nworkflow does not build or download the index.
--blacklist): defaults to the ENCODE Blacklist v2 URL for --genome.There is no sample sheet and no input control. Inputs are globs, and every sample in a run\nshares one --genome. Unlike ChIP-seq, ATAC-seq does not need a separate input or IgG\ncontrol; MACS2 calls peaks against a local background model.
The Tn5 transposase inserts sequencing adapters with a 9-bp duplication. To center\nreads on the actual cut site:
\nThe workflow applies this with alignmentSieve --ATACshift (deeptools) after duplicate\nremoval and before blacklist filtering. The correction is essential for footprinting and\nmotif analysis.
ATAC-seq produces a characteristic nucleosomal ladder pattern:
\n| Fragment Class | \nSize Range | \nBiological Meaning | \n
|---|---|---|
| Nucleosome-free (NFR) | \n<150 bp | \nOpen chromatin / TF binding | \n
| Mono-nucleosome | \n150-300 bp | \nSingle nucleosome wrapping | \n
| Di-nucleosome | \n300-500 bp | \nTwo nucleosomes | \n
| Tri-nucleosome | \n500-700 bp | \nThree nucleosomes | \n
The workflow calls peaks on the nucleosome-free BAM. The NFR/mono-nucleosome boundary is\n--nfr_max (default 150); the mono-nucleosome selection is --nfr_max to 300 bp. The\nworkflow does not plot the fragment size distribution.
main.nf)| Parameter | \nDefault | \nDescription | \n
|---|---|---|
--reads | \nnone (required) | \nGlob for the paired-end FASTQ file pairs | \n
--bowtie2_index | \n./<genome>_bowtie2_index | \nDirectory holding the Bowtie2 index files named <genome>.*.bt2 | \n
--genome | \nGRCh38 | \nGRCh38 or mm10; sets the MACS2 genome size, the default index directory and the default blacklist | \n
--blacklist | \nENCODE Blacklist v2 URL for --genome | \nBED (or .bed.gz) of artifact regions removed from the BAM | \n
--mito_name | \nchrM | \nName of the mitochondrial contig to drop | \n
--nfr_max | \n150 | \nMaximum nucleosome-free fragment length, and the lower bound of the mono-nucleosome selection | \n
--skip_idr | \nfalse | \nSkip the IDR step | \n
--single_end | \nfalse | \nAccepted but always rejected: the workflow stops with an error because it is paired-end only | \n
--outdir | \nresults | \nWhere results are published | \n
nextflow.config)| Parameter | \nDefault | \nDescription | \n
|---|---|---|
--container | \nencode-toolkit/pipeline-atacseq:1.0.0 | \nImage built from scripts/Dockerfile. Pass a registry image for gcp/aws, or a .sif file for slurm | \n
--max_cpus, --max_memory, --max_time | \n16, 64.GB, 24.h | \nUpper bounds applied to every process | \n
--slurm_queue, --slurm_account | \nnormal, none | \nSLURM partition and account | \n
--gcp_project, --gcp_workdir | \nnone (both required for -profile gcp) | \nGoogle Cloud project and gs:// work directory | \n
--gcp_location, --gcp_disk | \nus-central1, 200.GB | \nGoogle Batch region and per-task disk | \n
--aws_queue, --aws_workdir | \nnone (both required for -profile aws) | \nAWS Batch job queue and s3:// work directory | \n
--aws_region, --aws_cli_path | \nus-east-1, /home/ec2-user/miniconda/bin/aws | \nAWS region, and the AWS CLI path inside the Batch AMI | \n
Profiles are local, slurm, gcp and aws.
The workflow computes only the metrics marked \"workflow\" below. TSS enrichment,\nNRF/PBC, fragment-size plots and ataqv are manual post-processing steps documented in\nreferences/05-qc-metrics.md.
| Metric | \nThreshold | \nComputed by | \nSource | \n
|---|---|---|---|
| Total sequenced reads | \n>=50M (recommended) | \nworkflow (FastQC, flagstat) | \nENCODE | \n
| Mapping rate | \n>80% | \nworkflow (bowtie2 log, samtools flagstat) | \nENCODE | \n
| Mitochondrial fraction | \n<20% (ideal <5%) | \nworkflow (qc/<sample>.idxstats.txt) | \nENCODE | \n
| Duplication rate | \n<30% | \nworkflow (Picard dup_metrics.txt) | \nENCODE | \n
| IDR peaks at 0.05 | \n>50,000 | \nworkflow (peaks/idr/<sampleA>_vs_<sampleB>.idr_peaks.txt) | \nENCODE | \n
| NRF (non-redundant fraction) | \n>=0.8 | \nmanual | \nENCODE | \n
| PBC1 | \n>=0.8 | \nmanual | \nENCODE | \n
| TSS enrichment score | \n>=5 (GRCh38), >=6 (hg19), >=10 (mm10) | \nmanual (deeptools + a TSS BED) | \nENCODE standard | \n
| FRiP | \n>=0.3 | \nworkflow (qc/<sample>.frip_mqc.tsv) | \nENCODE | \n
| NFR fraction | \n>0.4 of fragments <150bp | \nmanual | \nBuenrostro 2013 | \n
qc/<sample>.idxstats.txt is samtools idxstats of the BAM before mitochondrial reads are\nremoved (contig, length, mapped, unmapped): the mitochondrial fraction is the mapped count\non the --mito_name row divided by the sum of the mapped column. MultiQC's samtools module\nreads the same file and reports that fraction.
The TSS enrichment score measures the fold enrichment of ATAC-seq signal at\ntranscription start sites compared to flanking regions. It is the single most\ninformative QC metric for ATAC-seq, but this workflow does not compute it: there is no\nTSS BED input and no computeMatrix/plotProfile step. Run it manually against\nsignal/<sample>.signal.bw with a TSS BED for your assembly; the commands are in\nreferences/05-qc-metrics.md.
| Score | \nQuality | \nInterpretation | \n
|---|---|---|
| >=7 | \nExcellent | \nHigh signal-to-noise | \n
| 5-7 | \nGood | \nAcceptable for most analyses | \n
| 3-5 | \nMarginal | \nReview other metrics carefully | \n
| <3 | \nPoor | \nLikely failed; consider re-doing | \n
nextflow run scripts/main.nf \\\n -profile local \\\n --reads 'fastq/*_R{1,2}.fq.gz' \\\n --genome GRCh38 \\\n --bowtie2_index GRCh38_bowtie2_index \\\n --blacklist hg38-blacklist.v2.bed.gz \\\n --outdir results/\n--blacklist is optional; without it the workflow downloads the ENCODE Blacklist v2 for\n--genome. Give the glob at least two replicates if you want the IDR step to run.
The slurm profile runs through Singularity, so pass a local image file rather than the\ndefault Docker image name:
singularity build pipeline-atacseq.sif docker-daemon://encode-toolkit/pipeline-atacseq:1.0.0\n\nnextflow run scripts/main.nf \\\n -profile slurm \\\n --container /path/to/pipeline-atacseq.sif \\\n --slurm_queue normal \\\n --reads 'fastq/*_R{1,2}.fq.gz' \\\n --genome GRCh38 \\\n --bowtie2_index GRCh38_bowtie2_index \\\n --outdir results/\n# Google Cloud Batch\nnextflow run scripts/main.nf -profile gcp \\\n --container us-docker.pkg.dev/<project>/<repo>/pipeline-atacseq:1.0.0 \\\n --gcp_project <project> \\\n --gcp_workdir gs://<bucket>/work \\\n --reads 'gs://<bucket>/fastq/*_R{1,2}.fq.gz' \\\n --genome GRCh38 \\\n --bowtie2_index gs://<bucket>/reference/GRCh38_bowtie2_index \\\n --outdir gs://<bucket>/results\n\n# AWS Batch\nnextflow run scripts/main.nf -profile aws \\\n --container <account>.dkr.ecr.<region>.amazonaws.com/pipeline-atacseq:1.0.0 \\\n --aws_queue <job-queue> \\\n --aws_workdir s3://<bucket>/work \\\n --reads 's3://<bucket>/fastq/*_R{1,2}.fq.gz' \\\n --genome GRCh38 \\\n --bowtie2_index s3://<bucket>/reference/GRCh38_bowtie2_index \\\n --outdir s3://<bucket>/results\n--outdir only sets where results are published; Google Batch and AWS Batch stage every\ntask through the work directory, and the workflow stops with an error if it or the\nproject/queue is missing.
| Platform | \nInstance | \nCost/Sample | \nTime/Sample | \nNotes | \n
|---|---|---|---|---|
| GCP | \nn1-standard-8 | \n~$2-4 | \n2-3 hours | \nSpot VMs enabled in the gcp profile | \n
| AWS | \nm5.2xlarge | \n~$2-4 | \n2-3 hours | \nSpot instances recommended | \n
| Local | \n8 cores, 32GB | \n$0 | \n3-5 hours | \nDocker required | \n
| SLURM | \n8 cores, 32GB | \nVaries | \n2-3 hours | \nSingularity image required | \n
results/\n fastqc/ # FastQC reports for raw and trimmed reads (.html, .zip)\n trimmed/ # Trimmed FASTQ (*_val_1.fq.gz / *_val_2.fq.gz) + trimming reports\n aligned/ # <sample>.bam, .bam.bai, <sample>.flagstat.txt,\n # <sample>.bowtie2.log\n filtered/ # <sample>.dup_metrics.txt, <sample>.final.bam(.bai),\n # <sample>.final.flagstat.txt\n shifted/ # <sample>.shifted.bam(.bai) -- Tn5-corrected, pre-blacklist\n nfr/ # <sample>.nfr.bam(.bai) and <sample>.mononuc.bam\n peaks/\n narrow/ # <sample>_peaks.narrowPeak, _summits.bed, _peaks.xls,\n # _treat_pileup.bdg, _control_lambda.bdg\n idr/ # <sampleA>_vs_<sampleB>.idr_peaks.txt (+ .png), one file per\n # replicate pair\n signal/ # <sample>.signal.bw (all fragments, RPKM)\n qc/\n <sample>.idxstats.txt # samtools idxstats before chrM removal:\n # contig / length / mapped / unmapped\n <sample>.frip_mqc.tsv # Peak set / FRiP / reads_in_peaks / total_reads\n multiqc/ # multiqc_report.html, multiqc_data/\n pipeline_info/ # timeline.html, report.html, trace.txt\nThe nucleosome-free and mono-nucleosome BAMs are both written to filtered/nfr/; there is\nno filtered/mononuc/ directory, and mononuc.bam has no index.
Mitochondrial DNA lacks chromatin and is highly accessible, often capturing 30-80%\nof reads. This is the most common ATAC-seq quality issue. The workflow removes --mito_name\nreads and publishes the per-contig counts it used, qc/<sample>.idxstats.txt: divide the\nmapped count on the --mito_name row by the sum of the mapped column to get the fraction,\nor read it from the samtools section of the MultiQC report. If >50% mito, consider\noptimizing the cell lysis step.
--mito_name defaults to chrM. Assemblies that call the contig MT need\n--mito_name MT. Confirm the name with samtools idxstats on a BAM from your index\nbefore running.
Bowtie2 handles the short fragments from ATAC-seq (especially NFR <150bp) better\nthan BWA-MEM. The workflow uses Bowtie2 with --very-sensitive.
--reads globIDR needs a pair of samples. With one sample there is no pair, so IDR is skipped silently\nand peaks/idr/ is never created. Every sample matched by --reads is treated as a\nreplicate of the same experiment, so unrelated samples in one glob produce meaningless\npairwise comparisons.
TSS enrichment is the most informative single metric for ATAC-seq, but the workflow does\nnot compute it. Run the manual computeMatrix/plotProfile step in\nreferences/05-qc-metrics.md before judging a library.
| File | \nDescription | \n
|---|---|
scripts/main.nf | \nNextflow DSL2 pipeline | \n
scripts/nextflow.config | \nExecution profiles (local/slurm/gcp/aws) | \n
scripts/Dockerfile | \nDocker image with all pipeline tools | \n
The image is pinned to linux/amd64; on an arm64 host it runs under emulation.
Tool versions in the image: Bowtie2 2.5.4, samtools 1.19, bedtools 2.31.0, Picard 3.1.1\n(Java 17), Trim Galore 0.6.10 with cutadapt 4.6, FastQC 0.12.1, MACS2 2.2.9.1,\nIDR 2.0.4.2, deepTools 3.5.5, MultiQC 1.21. The conda environment\nbioinformatics-installer/environments/atacseq-env.yml pins the same versions of the\ntools it lists.
After running on your own data, compare with ENCODE reference:
\n# Find matching ENCODE ATAC-seq experiments\nencode_search_experiments(\n assay_title=\"ATAC-seq\",\n organ=\"pancreas\",\n biosample_type=\"tissue\"\n)\n\n# Download ENCODE peaks for comparison\nencode_batch_download(\n download_dir=\"/data/encode_reference/\",\n output_type=\"IDR thresholded peaks\",\n assay_title=\"ATAC-seq\",\n organ=\"pancreas\",\n assembly=\"GRCh38\"\n)\nalignmentSieve --ATACshift after duplicate\nremoval. Without the correction, footprinting is offset by ~5 bp.bamPEFragmentSize manually.\n\n=10 (mm10) for ATAC-seq (ENCODE data standards). Values below 4 indicate poor\nsignal-to-noise. Computed manually, not by this workflow.
\n
--shift/--extsize do not apply here: the workflow calls peaks in -f BAMPE\nmode, where MACS2 takes fragment coordinates from read pairs, forces --nomodel and\nneutralises --shift internally. Shift/extension values only matter when calling peaks\non BED or single-end input. See references/04-peak-calling.md.--single_end is rejected with an error. Single-end ATAC-seq cannot\ndistinguish nucleosome-free from nucleosomal fragments.Goal: Process raw ATAC-seq FASTQ files through this pipeline to generate\nnucleosome-free region peaks and a signal track.\nContext: Bowtie2 alignment, chrM removal, duplicate removal, Tn5 shift (+4/-5),\nblacklist filtering, NFR selection and MACS2 peak calling.
\nencode_get_experiment(accession=\"ENCSR637ENO\")\nExpected output (fields abridged):
\n{\n \"accession\": \"ENCSR637ENO\",\n \"assay_title\": \"ATAC-seq\",\n \"biosample_summary\": \"GM12878\",\n \"assembly\": [\"GRCh38\"],\n \"bio_replicate_count\": 2,\n \"tech_replicate_count\": 2,\n \"status\": \"released\"\n}\nencode_list_files(experiment_accession=\"ENCSR637ENO\", file_format=\"fastq\")\nExpected output (a JSON array of files; fields abridged):
\n[\n {\"accession\": \"ENCFF100ATQ\", \"file_format\": \"fastq\", \"output_type\": \"reads\", \"biological_replicates\": [1], \"file_size_human\": \"1.8 GB\"},\n {\"accession\": \"ENCFF101ATQ\", \"file_format\": \"fastq\", \"output_type\": \"reads\", \"biological_replicates\": [1], \"file_size_human\": \"1.9 GB\"},\n {\"accession\": \"ENCFF102ATQ\", \"file_format\": \"fastq\", \"output_type\": \"reads\", \"biological_replicates\": [2], \"file_size_human\": \"1.7 GB\"},\n {\"accession\": \"ENCFF103ATQ\", \"file_format\": \"fastq\", \"output_type\": \"reads\", \"biological_replicates\": [2], \"file_size_human\": \"1.8 GB\"}\n]\nThe listing does not say which file of a pair is read 1 and which is read 2 -- no\nencode_* tool reports that. Open each file's page on encodeproject.org, where\npaired_end is 1 or 2 and paired_with names the other accession. Both replicates are\nneeded for the IDR step.
encode_download_files(file_accessions=[\"ENCFF100ATQ\", \"ENCFF101ATQ\", \"ENCFF102ATQ\", \"ENCFF103ATQ\"], download_dir=\"/data/atacseq/fastq\")\nENCODE FASTQs are named by accession (ENCFF123ABC.fastq.gz) with no _R1/_R2 in the\nname, so the --reads glob (*_R{1,2}.fq.gz) cannot pair them. Take the mate assignment\nfrom each file's page on encodeproject.org (paired_end is 1 or 2, paired_with names the\nother accession), then link them into the shape the glob expects:
cd /data/atacseq/fastq\nln -s ENCFF100ATQ.fastq.gz gm12878_rep1_R1.fq.gz\nln -s ENCFF101ATQ.fastq.gz gm12878_rep1_R2.fq.gz\nln -s ENCFF102ATQ.fastq.gz gm12878_rep2_R1.fq.gz\nln -s ENCFF103ATQ.fastq.gz gm12878_rep2_R2.fq.gz\nnextflow run scripts/main.nf \\\n -profile local \\\n --reads '/data/atacseq/fastq/gm12878_*_R{1,2}.fq.gz' \\\n --genome GRCh38 \\\n --bowtie2_index /data/reference/GRCh38_bowtie2_index \\\n --blacklist /data/reference/hg38-blacklist.v2.bed.gz \\\n --mito_name chrM \\\n --outdir /data/atacseq/results\nPipeline steps, in the order the workflow runs them:
\n--nextera), plus FastQC on the trimmed reads--very-sensitive, MAPQ 30, properly paired only)samtools idxstats of the pre-removal BAM published as qc/<sample>.idxstats.txtREMOVE_DUPLICATES=true)alignmentSieve --ATACshift)-f BAMPE --nomodel --keep-dup all --call-summits --qvalue 0.05 -B)From the workflow:
\n| Output | \nWhat to check | \n
|---|---|
qc/multiqc/multiqc_report.html | \nMapping rate (>80%), adapter content, duplication rate, mitochondrial fraction, FRiP | \n
qc/<sample>.idxstats.txt | \nMitochondrial fraction (<20%, ideal <5%): mapped reads on the chrM row over the sum of the mapped column | \n
qc/<sample>.frip_mqc.tsv | \nFRiP (>=0.3 for ATAC-seq) | \n
peaks/narrow/<sample>_peaks.narrowPeak | \nPeak count per replicate | \n
peaks/idr/gm12878_rep1_vs_gm12878_rep2.idr_peaks.txt | \nIDR peaks at 0.05 (>50,000) | \n
With the two replicates above there is one IDR file; a third replicate would add\ngm12878_rep1_vs_gm12878_rep3 and gm12878_rep2_vs_gm12878_rep3.
Manual follow-ups (not run by this workflow): TSS enrichment, fragment-size distribution\nplots, NRF/PBC and ataqv. Commands are in references/05-qc-metrics.md.
encode_track_experiment(accession=\"ENCSR637ENO\", notes=\"GM12878 ATAC-seq processed through the pipeline-atacseq skill\")\nencode_search_experiments(\n assay_title=\"ATAC-seq\",\n organ=\"pancreas\"\n)\nExpected output:
\n{\n \"results\": [\n {\n \"accession\": \"ENCSR789PAN\",\n \"assay_title\": \"ATAC-seq\",\n \"biosample_summary\": \"pancreas tissue male adult (44 years)\",\n \"status\": \"released\"\n }\n ],\n \"total\": 8,\n \"limit\": 25,\n \"offset\": 0,\n \"has_more\": false,\n \"next_offset\": null\n}\nencode_list_files(\n experiment_accession=\"ENCSR789PAN\",\n file_format=\"fastq\"\n)\nExpected output:
\n[\n {\n \"accession\": \"ENCFF100ATQ\",\n \"file_format\": \"fastq\",\n \"output_type\": \"reads\",\n \"biological_replicates\": [1],\n \"file_size_human\": \"3.1 GB\",\n \"status\": \"released\"\n }\n]\n| This skill produces... | \nFeed into... | \nPurpose | \n
|---|---|---|
| Accessible chromatin peaks | \naccessibility-aggregation | \nCross-experiment union merge | \n
| Peak regions (BED) | \nmotif-analysis | \nTF motif enrichment in open chromatin | \n
| Signal tracks (bigWig) | \nvisualization-workflow | \nGenome browser accessibility display | \n
| Nucleosome-free peaks | \nregulatory-elements | \nClassify accessible regions as enhancers/promoters | \n
| Peak coordinates | \nvariant-annotation | \nIdentify variants in accessible chromatin | \n
QC outputs (idxstats.txt, frip_mqc.tsv, MultiQC) | \nquality-assessment | \nValidate against ENCODE ATAC-seq standards | \n
pipeline_info/ reports | \ndata-provenance | \nRecord Tn5 shift, fragment filters, tool versions | \n
| Peak files | \njaspar-motifs | \nScan accessible regions for known TF motifs | \n
When reporting ATAC-seq pipeline results:
\nqc/<sample>.idxstats.txt -- mapped reads on\nthe --mito_name row over the sum of the mapped column -- or from the samtools section\nof the MultiQC report (ideal <5%, acceptable <20%)samtools flagstat),\nduplication rate (Picard dup_metrics.txt), and read counts, all aggregated in\nqc/multiqc/multiqc_report.htmlqc/<sample>.frip_mqc.tsv (>=0.3 for ATAC-seq). It is\ncomputed from the blacklist-filtered BAM (all fragments) against the peaks called on the\nnucleosome-free fragmentspeaks/narrow/, peaks/idr/, signal/,\nfiltered/nfr/, qc/, pipeline_info/)motif-analysis for TF footprinting and de novo motif discovery,\nor visualization-workflow for genome browser session generation