] simulates diffusions of the gene expressions in the spatial domain and models the expression diffusion with Fick’s second law to measure the time of convergence. In this context,] assumes that genes with spatial patterns will demonstrate a lower degree of randomness and a higher degree of structure.
SpaGCN [] is a graph convolutional network approach that integrates gene expression data, spatial location information, and histology images to identify genes with spatial patterns. The core of GCN is its graph convolutional layer, which enables it to combine graph structure and node information as inputs to a convolutional network.