Figure 2.

Schematic layout of the datasets used for SVM training and testing. The primary dataset consist of non-redundant tetrapeptide caspase substrate cleavage sites obtained from literature (see 1) and an equal number of non-cleavage sites. ¹The P₄P₁ sequences consist of all the sequences in the primary tetrapeptide cleavage site dataset. P₄P₂' and P₁₄ P₁₀' datasets were derived by extracting subsequence segments from the parent protein chains in the vicinity of the tetrapeptide cleavage sites, as shown in Figure 1. All datasets contain equal number of positive and negative examples.