Downloads

H5mu Files

H5mu files are specialized HDF5 files designed for efficient storage of single-cell multimodal data. They provide a structured and scalable format for handling complex datasets that include multiple data modalities, such as gene expression, cell surface protein abundance, and chromatin accessibility, all within a unified framework.

These files can be easily converted into single-modal formats depending on your analytical environment: In Python, you can transform h5mu files into h5ad files, which are widely used in the Scanpy ecosystem. In R, they can be converted into Seurat objects, making them compatible with the Seurat package for downstream analyses.

For CITE-seq datasets, each h5mu file contains three modalities: the raw gene expression modality, the cell surface protein modality, and the highly variable gene modality.

For 10x Genomics Multiome datasets, each h5mu file contains two modalities: the gene expression modality (including highly variable genes), and the chromatin accessibility modality (including highly variable peaks) .

CellChat RDS Files

These RDS files containing CellChat analysis data. You can easily download and import them into your local machine's R and conduct customized analyses.

Customized CellTypist Models

ScMultiModalMap employs CellTypist for performing cell type annotations. Beyond utilizing the official CellTypist models, scMultiModalMap has trained several customized models based on publicly available atlas-level single-cell datasets. These customized models have shown improved performance in cell type annotation tasks in scMultiModalMap and are available for download to facilitate the analysis of your own single-cell data.

If you have any question or comment, please feel free to contact us by scmmm@cncb.ac.cn.