Fig. 7: Cytarabine DRUML model predicts prognosis of AML patients treated with cytarabine.
From: Drug ranking using machine learning systematically predicts the efficacy of anti-cancer drugs
DRUML was used to predict responses to cytarabine from phosphoproteomics data obtained from 36 AML patients in triplicate by Casado et al. (PRIDE PXD005978). Prediction was obtained from the average prediction derived from random forest, principal component regression and partial least squares models. a, b Correlation between overall patient survival and cytarabine D values (a) or DRUML predicted responses (b) in patients that underwent complete remission (CR) and received consolidation therapy (p values by Spearman, n = 25). c, d Kaplan–Meier survival curves of patients with high (blue lines) or low (red lines) DRUML predicted cytarabine responses for CR (c) and for all (d) patients (p values by log-rank test). Mean of predicted cytarabine AAC was used as cut-off for patient stratification.
