Bulk ATAC-seq workflow

bulk-footprinting runs a complete bulk ATAC-seq analysis from prepared BAM/BAI and peak BED files. It produces footprint scores, differential motif results, and an interactive report.

This example compares three HepG2 and three K562 biological replicates from ENCODE experiments ENCSR291GJU and ENCSR868FGK.

Run from aligned files

Download the ENCODE BAM and peak sample sheet and its download helper. For your own data, use the local BAM and peak template. The same comparison file works for these aligned inputs.

bulk-footprinting --sample-table encode_hepg2_k562_bams.tsv \
  --comparison-table encode_hepg2_k562_comparisons.tsv \
  --genome hg38.fa.gz --blacklist hg38.blacklist.bed \
  --motif-db jaspar2026_vertebrates --outdir project --cores 8

Repeated condition names in a sample sheet define biological replicates. BAM, peak, blacklist, and genome files must use the same genome assembly.

Optional FASTQ-to-BAM preparation

On Linux, prepare-atac can prepare FASTQ files as BAM/BAI and peak BED inputs. Run it separately before bulk-footprinting:

prepare-atac --samples reads.tsv --genome hg38 --outdir prepared_project

Use the generated metadata/samples.tsv with bulk-footprinting. This preprocessing step is not part of the bulk footprinting wrapper.

Review the results

Open the generated interactive report to review differential motifs and aggregate footprint profiles. See the bulk output example or explore the ENCODE cancer-cell-line output demo.

For additional options and output paths, see the bulk-footprinting guide or the complete API reference.