De novo motif discovery

This optional workflow discovers enriched sequence motifs from candidate footprint intervals and compares them with known motif databases.

Main commands

call-footprints → discover-motifs → summarize-motifs

First use call-footprints with --output-bed to create candidate footprint intervals. Supply that BED file and a FASTA reference with the same assembly and chromosome names:

discover-motifs \
  --candidates candidate_footprints.bed \
  --genome hg38.fa.gz \
  --flank 75 \
  --method streme \
  --known-motif-db jaspar2026_vertebrates \
  --outdir project/de_novo \
  --execute

The command extracts sequences around the candidates, runs STREME, compares the discovered motifs with JASPAR using Tomtom, and writes a summary. External tools are installed automatically on first use. Without --execute, it only prepares the sequence file and a script for later execution.

Open project/de_novo/motif_summary.html to review the motifs and their known matches, or use motif_summary.tsv for further analysis. Similar TFs can share a motif, so a database match suggests a motif identity rather than confirming which protein binds. Use summarize-motifs to rebuild a summary from existing MEME, STREME, DREME, or Tomtom results.