Welcome to the molecular common coordinate framework (molCCF)!

How is the molCCF built

• We introduce FuseMap, a deep-learning-based framework for spatial transcriptomics that bridges single-cell or single-spot gene expression with spatial contexts and consolidates various gene panels across spatial transcriptomics atlases.
• We trained FuseMap on an extensive spatial transcriptomics atlas corpus of the mouse brain, comprising over 18.6 million cells or spots and 26,665 genes from 434 tissue sections across seven datasets in a self-supervised way, FuseMap gained a fundamental understanding of cell identities and gene characteristics. We thus achieved multiple tasks including nomenclature harmonization of molecular cell types and tissue regions, identification of novel molecular brain regions, spatial gene imputation, targeted gene-panel selection, and region-specific cell-type interactions inference.
• Based on this, we registered 434 coronal and sagittal spatial transcriptomics tissue sections from ≥11 animals within the Allen Mouse Brain Common Coordinate Framework Version 3, and achieved a comprehensive molecularly defined CCF (molCCF) that offers unified annotations of molecular cells and tissues with their corresponding anatomic location in the anaCCFv3.
• For more information, checkout our paper: Towards a universal spatial molecular atlas of the mouse brain!

What is in the Data Portal

This database includes spatially resolved single cells or spots across 434 sections, including 405 coronal sections and 29 sagittal sections from seven atlases. Each cell or spot has a 3D molCCF coordinate and is annotated:

• Universal molecular cell-type annotations encompassing 26 main types, 191 subtypes as per Atlas 1 Nomenclature
• Universal molecular tissue-region annotations with 17 main levels and 146 sublevel
• Imputed gene expression for 26,665 genes

Where to download the data

The following data will be available soon.

• Processed single-cell embedding in individual tissue slices (sample_processed_expression_pd.csv);
• Processed spatial embedding in individual tissue slices (sample_processed_expression_pd.csv);
• Processed gene embedding in individual tissue slices (sample_processed_expression_pd.csv);
• 3D molCCF coordinates and annotations for identified cells in individual tissue slices (sample_spatial.csv);
• Imputed single-cell gene expression of 26,665 genes in individual tissue slices (imputation_sample.h5ad).

How to cite

Yichun He, Hao Sheng, Hailing Shi, Wendy Xueyi Wang, Zefang Tang, Jia Liu, Xiao Wang. Towards a universal spatial molecular atlas of the mouse brain. Preprint at bioRxiv https://www.biorxiv.org/content/10.1101/2024.05.27.594872v1 (2024).

Please email Yichun He or Xiao wang with any questions, comments, or queries.

Acknowledgements

Publicly available datasets are used in this study (summarized in Supplementary Table 1). The following spatial transcriptomics atlases of mouse brain are used: Atlas 1, Atlas 2, Atlas 3, Atlas 4, Atlas 5, Atlas 6, Atlas 7. We thank the authors of these studies for making their data publicly available.

This site is managed by the Wang lab at the Broad Institute.