Machine Learning Assistants Construct Oxidative Stress-Related Gene Signature and Discover Potential Therapy Targets for Acute Myeloid Leukemia

Figure 1

Machine learning to identify important oxidative stress-related genes for the prognosis model. (a, b) The error rate of the random forest model is 29.83%, and the better and poor prognosis genes are ordered by their importance. (c) Thirty-four prognosis-related genes are important more than 0.3. (d, e) Fifteen candidate genes are extracted from the expression profile by the lasso regression model. (f) Six potential genes were identified after the merge of the results of the above two different machine learning algorithms.

(a)

(b)

(c)

(d)

(e)

(f)