Search the catalog
Filter datasets by display ID, source ID, sample source, biological condition, or model-usage labels before downloading large files.
Data Access
Download scMORA data on demand
Full 10x Multiome datasets are distributed as paired RNA-ATAC MuData objects through Hugging Face. Use the scmora-db Python package to search metadata, download selected datasets, and load files directly into analysis workflows.
Download selected data
Retrieve only the matched .h5mu files you need from the Hugging Face dataset repository.
Load into Python
Open downloaded MuData objects directly for downstream RNA, ATAC, and paired multiome analyses.
Python API
Search, download and load datasets
from scmora_db import search_datasets, download_datasets, load_datasets
catalog = search_datasets(
usage_tag="control",
detailed_condition="Control",
)
paths = download_datasets(
dataset_id="GM12878_01",
cache_dir="./scmora-cache",
)
mdata = load_datasets(
dataset_id="GM12878_01",
)
Command Line
Use scMORA from the terminal
scmora-db search --usage-tag control
scmora-db search --detail-source "GM12878 (Cell Line)"
scmora-db download --dataset-id GM12878_01
scmora-db load --dataset-id GM12878_01 --backed r
scmora-db list usage-tags
scmora-db list detailed-conditions
scmora-db list detail-sources
Repository Structure
What is provided
Paired multiome files
.h5mu objects preserve barcode-matched RNA and ATAC measurements from the same cells.
Curated metadata
Catalog fields include display ID, source ID, GSE accession, sample source, condition, cell count, and model-usage labels.
Reusable access layer
The package provides the same search and download logic for scripts, notebooks, and reproducible pipelines.