HDMA

Supplementary tables

Machine-readable copies of the manuscript supplementary tables that the data download and analysis workflows depend on. See ../DATA.md for the download recipes that use these, and the hdma agent skill (../.claude/skills/hdma/references/data.md).

file table what it is
table_s14.tsv Table S14 map of every deposited data type → Zenodo record/DOI (the download index)
table_s2.tsv Table S2 per-cluster cell metadata and the cluster-naming map
table_s6.tsv Table S6 motif lexicon / compendium annotations

table_s14.tsv — Table S14 (data deposition index)

The authoritative map used by the download recipes. One row per deposition; most data types span several “Part” depositions.

column description
data_type kind of data (filter on this to select a product)
size download size of the deposition
depo_name human-readable Zenodo deposition name
record_id Zenodo record ID (feed to get_urls.py)
url concept DOI (resolves to the latest version)

table_s2.tsv — Table S2 (cell metadata / cluster map)

One row per cluster. The compartment (col 5) and Cluster_ChromBPNet (col 19) columns drive the cell-type download recipes.

column description
Cluster unique cluster id
organ organ name
organ_code abbreviated organ name
cluster_id cluster id within organ
compartment cellular compartment (epithelial/endothelial/immune/stromal)
L1_annot level 1 cell type annotation (n=203)
L2_annot level 2 cell type annotation (n=134)
L3_annot level 3 cell type annotation (n=37)
dend_order order of cluster from hierarchically clustering L1 cell types
ncell number of cells in cluster
median_numi median RNA number of UMIs per cell
median_ngene median RNA number of genes per cell
median_nfrags median ATAC number of fragments per cell
median_tss median ATAC TSS enrichment ratio per cell
median_frip median ATAC fraction of reads in peaks per cell
note notes used for cell type annotation
organ_color color for clusters based on organ
compartment_color color for clusters based on compartment
Cluster_ChromBPNet cluster id used in ChromBPNet models

table_s6.tsv — Table S6 (motif lexicon / compendium)

One row per compendium motif. merged_pattern maps a compendium motif back to its per-cell-type constituent motifs.

column description
idx_uniq unique index for the motif within the compendium
motif_name unique motif name (prefix = idx_uniq, suffix = annotation)
motif_name_safe computationally safe motif_name (no slashes or pipe)
class motif contribution to accessibility (Activating or Repressive)
total_instances_across_celltypes total motif instances across all 189 modeled cell types
cwm_logo_fwd filename of the forward CWM logo PNG (see Table S14)
cwm_logo_rev filename of the reverse CWM logo PNG (see Table S14)
query_consensus consensus sequence for the trimmed CWM
annotation granular motif annotation (<family>:<TF>#<index> or <family>:<subfamily>/<alt>#<index>, Vierstra v2.1 convention; composite constituents joined by underscores)
annotation_broad broad motif annotation used to group motifs in analyses/figures
category motif category (base, base_with_flanks, homocomposite, heterocomposite, partial, repeat, unresolved)
best_match_TF TF associated with the best-matching known motif in databases (see Methods)
best_match_motif best-matching known motif in databases
best_match_TOMTOM_qval TOMTOM q-value for similarity to the best match
best_match_TFs_in_family other TFs in the same family as best_match_TF (HOCOMOCO)
best_match_TFs_in_Vierstra_archetype other TFs whose motifs share the Vierstra v2.1 cluster of the best match
total_seqlets_across_celltypes total seqlets across all constituent motifs aggregated into this compendium motif
merged_pattern name of the per-cell-type motif cluster corresponding to this compendium motif (maps back to constituents)
n_constituent_motifs number of motifs aggregated into this compendium motif
n_constituent_celltypes number of cell types contributing at least one constituent motif
constituent_organs number of organs contributing at least one constituent motif
constituent_celltypes L2_annot of the cell types contributing constituent motifs