Data for: Multimodal spatial transcriptomics redefines cortical pathology in Huntington Disease
Data files
Sep 01, 2026 version files 3.53 GB
-
adata_AGG_CAG.h5ad
1.80 GB
-
adata_short_all.h5ad
1.43 GB
-
C4_ref.h5ad
298.80 MB
-
README.md
19.42 KB
Abstract
This dataset contains processed multimodal MERFISH data generated from human cortical tissue to investigate cortical pathology in Huntington disease. The AnnData objects (.h5ad) include gene expression matrices, cell-level metadata (e.g., anonymized patient ID, HD/control status, cortical layer, cell-type cluster annotations, CAG repeat length, smFISH-derived HTT expression levels, huntingtin aggregate presence and characteristics, lipofuscin granule counts, and mean γH2AX, H3K9me, and H3K27ac levels), spatial coordinates, and dimensionality-reduction embeddings. These processed data were used to generate the main and supplementary figures presented in the associated manuscript.
Dataset DOI: 10.5061/dryad.fj6q57485
This dataset is associated with the manuscript "Multimodal spatial transcriptomics redefines cortical pathology in Huntington Disease."
This directory contains three AnnData (.h5ad) files generated from the spatial transcriptomics data. Each file contains single-cell gene expression data together with cell-level metadata, spatial coordinates, and dimensionality-reduction/integration results.
The files contain processed single-cell expression matrices and associated cell-level annotations generated from MERFISH imaging data. They are intended to provide the processed data required to reproduce the analyses and figures presented in the associated manuscript.
Description of the data and file structure
Files and variables
File: C4_ref.h5ad
Description: Human control cortex reference dataset used for cell-type annotation and reference-based analyses.
AnnData dimensions: 29,767 cells × 1,128 genes
This file contains the C4 Control dataset with expression measurements for 1,128 targeted genes and cell-level annotations, aggregate measurements, CAG-related measurements, and spatial information.
X — Total-count normalized and log1p-transformed expression matrix used for downstream analyses.
obs — cell-level metadata
| Field | Description |
|---|---|
volm |
Cell volume |
n_genes_by_counts |
Number of genes detected per cell |
log1p_n_genes_by_counts |
Log-transformed number of detected genes |
total_counts |
Total number of detected transcripts/counts per cell |
log1p_total_counts |
Log-transformed total transcript counts |
leiden |
Leiden clustering assignment |
vol_nuc |
Nuclear volume |
sample |
Sample identifier |
HD |
Disease status (HD True or False) |
patient |
Patient/donor identifier |
CAG_level |
CAG repeat signal/brightness measurement obtained by smFISH and used to estimate somatic CAG repeat length |
HTT_number |
Number of HTT transcripts detected by smFISH |
batch |
Experimental batch |
cluster_names |
Cell-type/subtype annotation |
cluster_general |
Major cell-type annotation |
Agg |
Huntingtin inclusion status: 1 = cell containing an intranuclear huntingtin inclusion; 0 = cell closest to an extranuclear inclusion; NaN = no associated inclusion. |
Agg_bright |
Inclusion brightness/intensity measurement |
Agg_vol |
Inclusion volume |
Agg_x |
Inclusion x-coordinate |
Agg_y |
Inclusion y-coordinate |
Agg_z |
Inclusion z-coordinate |
Agg_dist |
Inclusion distance to closest LaminA |
obsm
| Field | Description |
|---|---|
X_pca |
Principal component representation |
X_pca_harmony |
Harmony-integrated principal component representation |
X_raw |
Raw expression counts |
X_spatial |
Spatial coordinates |
X_umap |
UMAP embedding |
uns
Contains clustering, color assignments, PCA, UMAP, neighbor graph, and dendrogram information, including:
cluster_general_colors, cluster_names_colors, dendrogram_leiden, leiden, leiden_colors, neighbors, pca, and umap.
varm
PCs contains the principal component loadings.
obsp
connectivities and distances contain the cell-cell neighbor graph and pairwise neighbor distances.
File: adata_short_all.h5ad
Description: MM-188 MERFISH data for 6 Controls and 8 HD cortices.
AnnData dimensions: 428,973 cells × 188 genes
This file contains the spatial transcriptomics data used for cell-type quantifications, comprising 428,973 cells profiled for a reduced panel of 188 targeted genes.
X — Total-count normalized and log1p-transformed expression matrix used for downstream analyses.
obs — cell-level metadata
| Field | Description |
|---|---|
volm |
Cell volume |
n_genes_by_counts |
Number of genes detected per cell |
log1p_n_genes_by_counts |
Log-transformed number of detected genes |
total_counts |
Total number of detected transcripts/counts per cell |
log1p_total_counts |
Log-transformed total transcript counts |
leiden |
Leiden clustering assignment |
sample |
Sample identifier |
HD |
Disease status (HD True or False) |
patient |
Patient/donor identifier |
batch |
Experimental batch |
cluster_names |
Cell-type/subtype annotation |
cluster_general |
Major cell-type annotation |
BBData |
Dataset designation: 'new' = extended dataset; 'old' = original dataset used throughout the manuscript |
obsm
| Field | Description |
|---|---|
X_pca |
Principal component representation |
X_raw |
Raw expression counts |
X_spatial |
Spatial coordinates |
X_umap |
UMAP embedding |
File: adata_AGG_CAG.h5ad
Description: MM-1128 MERFISH data for 4 Controls and 5 HD cortices.
AnnData dimensions: 185,721 cells × 1,128 genes
This file contains the subset of cells for which CAG repeat and/or aggregate-related measurements were available, with expression measurements for 1,128 targeted genes and additional measurements related to CAG repeat length, huntingtin aggregates, lipofuscin, and chromatin-associated fluorescence measurements.
X — Total-count normalized and log1p-transformed expression matrix used for downstream analyses.
obs — cell-level metadata
| Field | Description |
|---|---|
volm |
Cell volume |
n_genes_by_counts |
Number of genes detected per cell |
log1p_n_genes_by_counts |
Log-transformed number of detected genes |
total_counts |
Total number of detected transcripts/counts per cell |
log1p_total_counts |
Log-transformed total transcript counts |
leiden |
Leiden clustering assignment |
vol_nuc |
Nuclear volume |
sample |
Sample identifier |
HD |
Disease status (HD True or False) |
patient |
Patient/donor identifier |
CAG |
CAG repeat signal/brightness measurement obtained by smFISH and used to estimate somatic CAG repeat length |
nHTT |
Number of HTT transcripts detected by smFISH |
batch |
Experimental batch |
cluster_names |
Cell-type/subtype annotation |
cluster_general |
Major cell-type annotation |
Agg |
Huntingtin inclusion status: 1 = cell containing an intranuclear huntingtin inclusion; 0 = cell closest to an extranuclear inclusion; NaN = no associated inclusion |
Agg_bright |
Inclusion brightness/intensity measurement |
Agg_vol |
Inclusion volume |
Agg_x |
Inclusion x-coordinate |
Agg_y |
Inclusion y-coordinate |
Agg_z |
Inclusion z-coordinate |
Agg_dist |
Inclusion distance to closest LaminA |
lipo |
Lipofuscin granule numbers measured |
m_gamma |
Mean γH2AX fluorescence/intensity measurement |
v_gamma |
Volume of γH2AX measurement |
m_h3k9 |
Mean H3K9me fluorescence/intensity measurement |
m_h3k27 |
Mean H3K27ac fluorescence/intensity measurement |
obsm
| Field | Description |
|---|---|
X_pca |
Principal component representation |
X_pca_harmony |
Harmony-integrated principal component representation |
X_raw |
Raw expression counts |
X_spatial |
Spatial coordinates |
X_umap |
UMAP embedding |
Code and software
Analysis code for data processing, analysis, and visualization is available at:
https://github.com/oleolatz/HDMMERFISH
All three files are provided in AnnData (.h5ad) format and can be opened using the Python anndata and scanpy packages.
The expression matrix is stored in the AnnData object (X), cell-level annotations are stored in obs, and spatial coordinates and dimensionality-reduction representations are stored in obsm.
Expression preprocessing
X_raw contains the original expression values prior to normalization. X contains total-count normalized and log1p-transformed expression values used for downstream analyses.
Raw expression values were used to calculate per-cell total transcript counts and quality-control metrics. Counts were normalized using sc.pp.normalize_total, with a target sum equal to the median total transcript count across cells, followed by natural-log transformation using sc.pp.log1p.
Thus, for all three datasets:
- X_raw --> raw expression counts
- X --> total-count normalized and log1p-transformed expression values
The 1,128-gene files contain the larger targeted gene panel used for the main multimodal spatial transcriptomics analyses. The 188-gene dataset contains a reduced gene panel and was used for cell-type quantification across the expanded cohort.
Cell-type annotations are provided in cluster_names (cell-type/subtype level) and cluster_general (major cell-type level).
X_spatial contains the spatial coordinates of individual cells and can be used to reproduce spatial analyses and visualizations described in the associated study.
Citation
Please cite the associated publication when using these data.
Dataset DOI: https://doi.org/10.5061/dryad.fj6q57485
Human subjects data
All materials have been removed of any patient identifiers. All Material has been collected from donors for or from whom a written informed consent for a brain autopsy and the use of the material and clinical information for research purposes had been obtained by the NBB and UWA.