Skip to main content

Table 3 EvP model assessment accuracy for different selection protocols

From: Evolutionary potentials: structure specific knowledge-based potentials exploiting the evolutionary record of sequence homologs

Selection protocol

AUC (%)

ACC (%)

OT

FPR (%)

TPR (%)

EvP (CLUSTER)

99.5

97.4

-4.18

2.3

97.0

EvP (PSI-BLAST)

97.7

94.2

-1.86

2.2

89.2

EvP (BLAST)

97.5

93.5

-2.06

1.7

86.9

EvP (RND)

71.9

68.6

-1.82

12.1

42.3

  1. EvPs in this table were generated using the CLS_90-90 clustering cut-offs and the MSA_20 MSA filtering cut-off. The sets of EvPs are named according to the clustering parameters (CLS_XX-YY) and the multiple sequence alignment filters (MSA_ZZ) used to derive them (Materials and methods). XX is the minimal structural similarity and YY is the minimal sequence identity used for structural clustering. ZZ is the minimal sequence identity shared between a sequence in the multiple sequence alignment and the sequence of the representative structure of the cluster. AUC is the area under the ROC curve, ACC is the maximal accuracy, OT is the optimal classification threshold, FPR is the false positive rate, and TPR is the true positive rate.