Skip to main content
Fig. 2 | BMC Medical Genomics

Fig. 2

From: Identification of contributing genes of Huntington’s disease by machine learning

Fig. 2

The decision tree model. a The program workflow. b The receiver operating characteristic (ROC) curve showing the performance of the prediction power of the model. c The modeled decision tree. A decision tree plots the “If… then” splitting of samples for prediction. The nodes denote the attributes, while the arrows denote the split, which meets a certain criterion. The number in the result box denotes the prediction result of the model (1 = HD; 0 = control), and the bar denotes the actual sample disease characteristic, bar sickness for sample size and bar segment for the proportion of HD samples (red = HD, blue = control). d The sample distribution in the 3-dimensional eigenspace of gene profiling. Red = HD, blue = control

Back to article page
\