Prediction of conformational epitopes with the use of a knowledge-based energy function and geometrically related neighboring residue characteristics

Table 2 Average performance of the CE-KEG for using average energy function of local neighboring residues.

Weighing Combinations	SE	SP	PPV	ACC
0%EG+100% GAAP	0.478	0.831	0.266	0.796
10%EG + 90% GAAP	0.490	0.831	0.273	0.797
20%EG + 80% GAAP	0.492	0.831	0.275	0.797
30%EG + 70% GAAP	0.497	0.831	0.277	0.798
40%EG + 60% GAAP	0.493	0.832	0.280	0.799
50%EG + 50% GAAP	0.503	0.834	0.284	0.801
60%EG + 40% GAAP	0.504	0.834	0.284	0.801
70%EG + 30% GAAP	0.519	0.839	0.294	0.808
80%EG + 20% GAAP	*0.531*	0.840	0.300	0.811
90%EG + 10% GAAP	0.521	0.839	0.294	0.809
100%EG + 0% GAAP	0.496	0.837	0.279	0.805

The performance used combinations of weighting coefficients for the average energy (EG) and frequency of geometrically related pairs of predicted CE residues (GAAP) within a 8-Å radius sphere. The highest SE is denoted by a bold-italic face.

ISSN: 1471-2105