Shaow/mousehypothalamus_merfish_moffitt

    # Mouse Hypothalamic Preoptic Region MERFISH · Moffitt 2018

    Curated, ready-to-load spatial transcriptomics dataset.

    ## Source

    - Paper: [Moffitt et al., Science 2018](https://www.science.org/doi/10.1126/science.aau5324)
    - Canonical download: Bioconductor MerfishData::MouseHypothalamusMoffitt2018() — mirrors Dryad doi:10.5061/dryad.8t8s248

    ## Scale

    | Property | Value |
    |---|---|
    | Technology | MERFISH (135 smFISH + 21 seqFISH = 156 genes) |
    | Species | Mus musculus |
    | Tissue | Mouse hypothalamic preoptic region |
    | Sections / slices | 12 |
    | Total cells / spots | 73,642 |

    ## Files

    - `Moffitt_hypothalamus_full.h5ad`

    Each `.h5ad` follows the AnnData spec:
    - `.X` — gene expression matrix (cells × genes), sparse where natural
    - `.obs` — per-cell annotations (see "Metadata" below)
    - `.obsm['spatial']` — `(n_cells, 2)` float32 spatial coordinates
    - (where present) `.layers['count']` — raw integer counts
    - (where present) `.obsm['spatial3d']` — `(n_cells, 3)` float32 (x, y, z=section)

    ## Metadata (`obs` columns)

    `cell_class`, `neuron_cluster_id`, `sample_id`, `behavior`, `sex`

    ## Notes

    Moffitt 2018 MERFISH hypothalamic-preoptic-region atlas, Animal 1 only (73,642 cells × 156 genes across 12 sagittal sections at Bregma -0.29..0.26). The smFISH primary panel (135 genes) and seqFISH altExp (21 genes) are concatenated. `obsm['spatial']` holds (x,y); `obsm['spatial3d']` holds (x,y,z=Bregma) for 3D registration tasks.

    ## Usage

    ```python

import scanpy as sc, numpy as np from huggingface_hub import snapshot_download d = snapshot_download(repo_id='Shaow/mousehypothalamus_merfish_moffitt', repo_type='dataset') a = sc.read_h5ad(f'{d}/Moffitt_hypothalamus_full.h5ad')

filter to a Bregma slice if desired

z = a.obsm['spatial3d'][:, 2] sub = a[(np.isclose(z, -0.09)) & (a.obs['cell_class'].astype(str) != 'Ambiguous')]

        ## Citation

        If you use this dataset, please cite the source paper above.

        ## License

        MIT for the curation/preparation. Underlying data inherits the license of
        the upstream publication (typically CC-BY-4.0); please see the source paper.