Produce a SummarizedExperiment instance with assay elements (i,j) the simba score relating gene i to cell j

embs_to_scores(embout, sce, colfeat = "Barcode", rowfeat = "Symbol")

Arguments

embout: output of `sce_to_embeddings`
colfeat: character(1) element of colData of `sce` to index cells
rowfeat: character(1) element of rowData of `sce` to index genes

Value

a SummarizedExperiment

Note

The simba algorithm produces probability-like scores for measuring tendency of two entities to be linked. This function takes the embeddings produced by sce_to_embeddings and produces edgewise softmax rescalings of C'G where C is the embedding for cells and G the embedding for genes