Figure 1.

The procedure of generating KNN graph with K = 2. Each column of the similarity matrix represents the similarity between one data point and others, in which two shaded units are the 2 nearest neighbors of one data point. The adjacency matrix is derived from the similarity matrix.

Li and Li BMC Genomics 2012 13(Suppl 7):S27   doi:10.1186/1471-2164-13-S7-S27