Global consensus map of human transcription factor footprints

Vierstra, J. et al. Global reference mapping of human transcription factor footprints. Nature 583, 729–736 (2020). https://doi.org/10.1038/s41586-020-2528-x

Preprint @ bioRxiv: https://doi.org/10.1101/2020.01.31.927798

Contact: Jeff Vierstra ([e-mail])

Genomic DNase I footprinting enables quantitative, nucleotide-resolution delineation of sites of transcription factor occupancy within native chromatin. We combined sampling of >67 billion uniquely mapping DNase I cleavages from >240 human cell types and states to index, with unprecedented accuracy and resolution, human genomic footprints and thereby the sequence elements that encode transcription factor recognition sites.

Please see http://vierstra.org/resources/dgf for additional information and a complete set of raw DNase I data for individual datasets. Additionally, raw data can also be accessed via the ENCODE data portal (http://encodeproject.org) using the dataset accessions found in Supplementary Table 1.

Code for footprint analysis and tutorials on how to access and manipulate digital genomic footprint data can be found at https://footprint-tools.readthedocs.io/en/latest/.

All files herein correspond to human genome build version GRCh38 (UCSC hg38).

Dataset contents:

Biosample metadata – Supplementary_Table_1.xlsx

Motif clustering metadata – Supplementary_Table_2.xlsx

ChIP-seq validation metadata – Supplementary_Table_3.xlsx

Consensus footprint coordinates and assigned motif archetypes
TSV file (BED-format) with consensus footprint (posterior probability>0.99) coordinates and overlaps with matches to motif model clusters. The legend file contains column definitions in detail.

consensus_footprints_and_motifs_hg38.bed.gz

consensus_footprints_and_motifs_legend.txt

Motif archetype matches overlapping consensus footprints
TSV file (BED-format) containing the coordinates for clustered motif model matches that overlap consensus footprints

collapsed_motifs_overlaping_consensus_footprints.bed.gz

collapsed_motifs_overlaping_consensus_footprints_legend.txt

Footprint occupancy matrix of consensus footprints
Rows are same order as the consensus footprint file and columns are same order as in the metadata files.

consensus_index_matrix_full_hg38.txt.gz (Values are –log(1-posterior))

consensus_index_matrix_binary_hg38.txt.gz (binary occupancy matrix, where footprints with posterior footprint probability >0.99 are considered occupied)

Single nucleotide variants tested for allelic imbalance 
The legend file contains column definitions in detail.

genotypes.vcf.gz - Genotyping and allelic read depth for each biosample (see header for more information)

tested_snvs_padj.bed.gz - SNVs tested for imbalance (TSV, BED-format)

tested_snvs_padj_legend.txt