Novel AI-based approach for cell type deconvolution in Visium datasets when reference scRNA-seqs include multiple minor cell types
Novel AI-based approach for cell type deconvolution in Visium datasets when reference scRNA-seqs include multiple minor cell types
Taguchi, Y.-h.; Turki, T.
AbstractWe have applied tensor decomposition (TD)-based unsupervised feature extraction (FE) to integrate multiple Visium datasets, as a platform for spatial gene expression profiling (spatial transcriptomics). As a result, TD-based unsupervised FE successfully obtains singular value vectors consistent with spatial distribution, i.e., singular value vectors with similar values are assigned to neighboring spots. Further, TD-based unsupervised FE successfully infers the cell type fractions within individual Visium spots (i.e., successful deconvolution) by referencing single-cell RNA-seq experiments that include multiple minor cell types, for which other conventional methods - RCTD, SPOTlight, SpaCET, and cell2location - fail. Therefore, TD-based unsupervised FE can be used to perform deconvolution even when other conventional methods fail as it includes multiple minor cell types in the reference profiles. TD-based unsupervised FE is thus expected to be applied to a wide range of deconvolution applications.