geo_small
收藏Zenodo2025-12-08 更新2026-05-26 收录
下载链接：
https://zenodo.org/doi/10.5281/zenodo.17717317
下载链接
链接失效反馈官方服务：
资源简介：
GEO dataset with 700k samples, subsetted to 350k samples. Annotated with natural language descriptions obtained in CellWhisperer Study.
## Processed AnnData Objects
This deposit contains processed AnnData objects (single-cell genomics data) that were processed using the `adata_hf_datasets` Python package. The files are in HDF5 (.h5ad) or Zarr format and contain preprocessed gene expression data along with computed embeddings.
### Available Embeddings (.obsm layers)
The AnnData objects contain the following embedding layers: `X_pca`, `X_scvi_fm`, `X_gs10k`, `X_geneformer`, `X_geneformer-v1`.
## Related Resources
- **HuggingFace Dataset**: [jo-mengr/geo_small_multiplets_natural_language_annotation](https://huggingface.co/datasets/jo-mengr/geo_small_multiplets_natural_language_annotation)- **Raw Data Source**: [Download](https://medical-epigenomics.org/papers/schaefer2024/data/datasets/archs4_geo/full_data.h5ad)
## Package Information
These files were processed using the `adata_hf_datasets` package. For more information about the processing pipeline, please refer to the associated HuggingFace dataset repository.
提供机构：
Zenodo创建时间：
2025-12-08
