A Novel Prognostic Model for Acute Myeloid Leukemia Based on Gene Set Variation Analysis

Figure 2

Identification of relevant subtypes of AML using immunologic and hallmark gene sets. (a) The optimal number of clusters (k) was determined with the factoextra package. (b) Visualization of cluster results, with k being 3. (c) The nonnegative matrix factorization (NMF) method was used to cluster AML samples. (d) The silhouette width plots displayed the clustering effect. The values of the silhouette range [−1, 1], where a high value indicates that the object is well matched to its own cluster and poorly matched to neighboring clusters. (e) Survival analysis among different subtypes.

(a)

(b)

(c)

(d)

(e)