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

Figure 6

Prognostic model construction and validation. (a) LASSO coefficients of prognostic gene sets. (b) Three-fold cross-validation for the selection of prognostic gene sets in LASSO regression. Prognostic model was constructed based on LASSO regression analysis and validated in TCGA (c), GSE146173 (d), and GSE12417 cohorts (e). Patients in TCGA (f), GSE146173 (g), and GSE12417 (h) cohorts were divided into low- and high-risk groups according to the ES of GSE35543_IN_VITRO_ITREG_VS_CONVERTED_EX_ITREG_DN with median ES as the cutoff value. Patients in TCGA (i), GSE146173 (j), and GSE12417 (k) cohorts were divided into low- and high-risk groups according to the ES of HALLMARK_CHOLESTEROL_HOMEOSTASIS with median ES as the cutoff value.

(a)

(b)

(c)

(d)

(e)

(f)

(g)

(h)

(i)

(j)

(k)