Skip to main content

Table 3 Comparison of prediction errors for different depth bins for the RDPred and the simulation of YW method on YW923 dataset.

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

  

PSSM+PS

RDPred

Depth bin

# of residues

MAE

MRE (%)

MAE

MRE (%)

  

0.25 (0.035,0.525)

14.8 (2.1,30.6)

0.19 (0.026,0.380)

11.0 (1.5,22.1)

1.75 ≤ RD < 2.0

56537

0.27 (0.034,0.604)

14.7 (1.8,32.6)

0.21 (0.021,0.474)

11.3 (1.2,25.5)

2.0 ≤ RD < 2.25

29172

0.33 (0.041,0.733)

15.6 (2.6,34.8)

0.29 (0.034,0.653)

13.7 (1.6,30.9)

2.25 ≤ RD < 2.5

15647

0.41 (0.061,0.845)

17.2 (2.6,35.7)

0.40 (0.062,0.794)

16.7 (2.6,33.4)

2.5 ≤ RD < 3.0

14203

0.51 (0.085,0.963)

18.6 (3.1,35.1)

0.53 (0.096,0.966)

19.6 (3.5,34.7)

3.0 ≤ RD < 4.0

12654

0.77 (0.153,1.490)

22.3 (4.5,40.2)

0.82 (0.170,1.495)

23.9 (5.0,41.8)

4.0 ≤ RD < 5.0

8311

1.32 (0.338,2.262)

29.1 (7.6,49.2)

1.33 (0.300,2.319)

29.4 (6.6,50.8)

5.0 ≤ RD < 6.0

6421

1.91 (0.643,3.041)

34.8 (11.8,55.3)

1.87 (0.497,3.127)

34.2 (9.0,57.1)

RD ≥ 6.0

6282

3.32 (1.419,5.35)

44.1 (21.0,64.7)

3.26 (1.158,5.386)

43.1 (17.1,65.8)

  1. The values in the brackets correspond to the 10th and 90th centiles of the absolute and relative errors.