Skip to main content
. 2021 Mar 17;22(6):3079. doi: 10.3390/ijms22063079

Figure 1.

Figure 1

Performance evaluation of the models using the 2-mer SeqPose features after filtering out some redundant features. The vertical axis gives the values of the five performance metrics, including Sn, Sp, MCC, AUC, and Acc. The horizontal axis is the number of features removed from the 2-mer SeqPose features, i.e., “−5” means that the five features with the largest p-values are removed from the list of all 2-mer SeqPose features. The feature removal is carried out in two steps, i.e., (a) five features are removed in each iteration; and then (b) one feature is removed in each iteration from the list of features “−45”.