Research Paper Volume 13, Issue 4 pp 5718—5747

Comprehensive analysis of macrophage-related multigene signature in the tumor microenvironment of head and neck squamous cancer

The running result of the random forest model. (A) The random forest model has the lowest error rate and tends to be stable after generating 1000 decision trees (ntree=1000). (B) The top five genes with highest feature important score were selected, namely FANCE, UTP3, DNAJC13, ADAMDEC1, and DNASE1L3.

Figure 3. The running result of the random forest model. (A) The random forest model has the lowest error rate and tends to be stable after generating 1000 decision trees (ntree=1000). (B) The top five genes with highest feature important score were selected, namely FANCE, UTP3, DNAJC13, ADAMDEC1, and DNASE1L3.