mims-harvard/SPATIA_MIST
MIST Dataset Construction For SPATIA (ICML 2026) MIST (Multimodal Imaging and Spatial Transcriptomics) is the pretraining dataset. It combines cell-level gene expression with morphology image crops from Xenium spatial transcriptomics data. Construction has three stages: Stage A: Crop cell images into LMDB Crop cell-centered patches from Xenium morphology TIFF images and store them in an LMDB database. Input layout (per Xenium dataset): dataset_dir/ ├──… See the full description on the dataset page: https://huggingface.co/datasets/mims-harvard/SPATIA_MIST.
| Author / organization | mims-harvard |
|---|---|
| License (as declared) | mit |
| Downloads | 492 (as reported by Hugging Face) |
| Likes | 0 |
| Created | 29 June 2026 |
| Last updated at source | 4 July 2026 |
Popularity, last 3 days
492 now (-105 since 2026-10-16). Counts as reported by the source, snapshotted daily by Silytics.
More from mims-harvard
- mims-harvard/ProCyon-InstructDataset
- mims-harvard/ToolSpaceDataset
- mims-harvard/MedeaDBDataset
Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 18 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.