Skip to main content

Table 2 Increase of the prediction performance upon sequential addition of features. The window size is W = 9, and the AUC score is evaluated in 10-fold cross validation on the S85 dataset

From: SEPIa, a knowledge-driven algorithm for predicting conformational B-cell epitopes from the amino acid sequence

Feature combination AUC score
F1 0.619
F1 + F2 0.624
F1 + F2 + F10 0.629
F1 + F2 + F10 + F11 0.630
F1 + F2 + F10 + F11 + F12 0.631
F1 + F2 + F9 + F10 + F11 + F12 0.631
F1 + F2 + F6 + F9 + F10 + F11 + F12 0.636
F1 + F2 + F3 + F6 + F9 + F10 + F11 + F12 0.636
F1 + F2 + F3 + F6 + F9 + F10 + F11 + F12 + F13 0.637
F1 + F2 + F3 + F6 + F9 + F10 + F11 + F12 + F13 + F7 0.640
F1 + F2 + F3 + F6 + F9 + F10 + F11 + F12 + F13 + F7 + F4 0.644
F1 + F2 + F3 + F6 + F9 + F10 + F11 + F12 + F13 + F7 + F4 + F5 0.644
F1 + F2 + F3 + F6 + F9 + F10 + F11 + F12 + F13 + F7 + F4 + F5 + F8 0.646
  1. The largest AUC score is indicated in bold
Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us