Skip to main content

Table 1 Performance evaluation for different combinations of features by 10-fold cross-validation using RevSS_PDB dataset

From: Prediction of reversible disulfide based on features from local structural signatures

Feature sets ACC SN SP MCC AUC
AAC + DST + LEN + PCP + SST 0.750 0.352 0.953 0.405 0.751
AAC + DST + PCP + SST 0.743 0.339 0.949 0.383 0.749
AAC + DST + LEN + PCP 0.738 0.291 0.967 0.375 0.740
AAC + DST + LEN + SST 0.728 0.283 0.956 0.341 0.726
AAC + LEN + PCP + SST 0.729 0.339 0.929 0.344 0.719
DST + LEN + PCP + SST 0.729 0.265 0.967 0.348 0.698
  1. The results were ordered by AUC values