In the core–periphery pattern, we can observe 4 blocks in the class blocked adjacency matrix, namely the core, core–periphery, periphery–core and periphery. Since nodes that are position equivalent would exhibit similar behaviors and outcomes, the partitioning of core nodes and periphery nodes ...
Due to the relatively low sequencing depth and limited coverage of single-cell RNA-seq data, the probabilities of many AS events cannot be reliably estimated for all the single cells in a dataset. Therefore, the AS probability matrix greatly suffers from frequent missing values (marked as NA)....
Next, the raw similarity matrix is used to generate a weighted adjacency matrix for a weighted, directed cell graph G, in which a heavy edge from cell i to cell j indicates strong attraction of the former to the latter. Then, the balanced similarity graph G is partitioned into metacell ...
To address this issue, we conceived of a simple approach in which for each gene only the edges reflecting the most highly correlated genes are retained (Edge Per Gene: EPG; where EPG3 = retaining 3 edges per gene). These are assembled into a matrix in which a gene may retain additi...