Fig. 1
From: Flexible analysis of spatial transcriptomics data (FAST): a deconvolution approach
Illustration of the FAST pipeline based on regularized non-negative matrix factorization. X is an N-by-M matrix that contains gene count data per spot, while W and H are a pair of low-rank embeddings of X. W addresses the signature genes of each cell, and H contains the cell proportions in each spot. FAST, as a unified model, deconvolves X into W and H, with histology image data embedded in the objective function