Skip to main content

Table 10 Comparison of residue depth prediction quality when removing individual sets of features used by RDPred.

From: Sequence based residue depth prediction using evolutionary information and predicted secondary structure

Feature sets used

PCC

MAE

MRE(%)

RDPred (all features)

0.668

0.558 (0.037,1.510)

17.0 (1.8,41.4)

PSSM removed

0.518

0.656 (0.041,1.907)

18.9 (2.1,47.4)

PS removed

0.632

0.575 (0.037,1.555)

17.3 (1.9,42.3)

SS removed

0.650

0.575 (0.037,1.556)

17.3 (1.9,42.4)

PI removed

0.664

0.562 (0.037,1.519)

17.1 (1.8,41.6)

  1. The tests were performed on the YW923 dataset using three-fold cross validation. The values in the brackets correspond to the 10th and 90th centiles of the absolute and relative errors.