single2spatial-spatial-mapping

To map scRNA-seq references to spatial transcriptomics, this Skill trains a deep-forest-based Single2Spatial model to align preprocessed, log-normalized scRNA-seq data with spatial coordinates, enabling spot-level cell-type proportion inference across tissues.

Spatial mapping requires preprocessed, log-normalized scRNA-seq data and spatial transcriptomics data with spatial coordinates stored in the spatial_data.obs attribute, typically structured within AnnData objects.

Yes, you can use optional GPU acceleration when training the deep-forest-based Single2Spatial model to align scRNA-seq references with spatial transcriptomics profiles for reconstructing tissue architecture.

Single2Spatial infers spot-level cell-type proportions by training a deep-forest mapper to translate scRNA-seq cell-type compositions into spatially resolved maps, visualizing marker expression on tissue coordinates.

The best way to reconstruct tissue architecture is training a deep-forest-based mapper on matched scRNA-seq and spatial transcriptomics data, exporting generated AnnData objects and summary reports for downstream analysis.

Log-normalized scRNA-seq data is required for spatial cell-type mapping because the deep-forest model relies on normalized gene expression to accurately translate single-cell references into spatially resolved cell-type distributions.