-Profiles: A Nonlinear Clustering Method for Pattern Detection in High Dimensional Data

Figure 3

An example of confusion matrices shown as images. Cleaner pictures indicate better agreement between true clusters and clustering results. The left-most column of each subplot represents the pure noise gene group. (a) -profiles clustering result. (b) GDHC result. (c) -means result.

(a)

(b)

(c)