Skip to main content

Table 2 Performance of the distance map algorithm

From: Toward an accurate prediction of inter-residue distances in proteins using 2D recursive neural networks

Seq. id [%]

[0, 10)

[10,20)

[20,30)

[30,40)

[40,50)

[50,60)

[60,70)

[70,80)

[80,90)

[90,95)

All

Model

           

TB

5.7±3.4

6.5±3.5

4.4±3.2

3.1±2.1

2.6±1.6

2.4±1.5

2.4±1.3

2.3±1.3

2.5±1.4

2.6±1.9

3.70±2.9

 

(7.1)

(7.1)

(4.6)

(3.3)

(3.1)

(2.5)

(2.5)

(2.3)

(2.4)

(2.8)

(4.52)

AI classical

5.5±2.5

6.3±2.8

5.9±2.5

5.8±2.6

5.6±1.9

5.7±2.0

5.5±1.9

5.6±2.0

6.0±2.9

6.1±3.2

5.85±2.6

 

(7.0)

(6.9)

(6.6)

(6.5)

(6.6)

(6.3)

(6.3)

(6.5)

(6.7)

(6.8)

(6.75)

AI

5.5±2.6

6.3±2.8

5.9±2.6

5.8±2.5

5.6±1.9

5.7±2.1

5.5±1.9

5.7±2.0

6.0±3.0

6.1±3.1

5.85±2.6

Compl.

(7.1)

(6.9)

(6.7)

(6.5)

(6.6)

(6.3)

(6.2)

(6.4)

(6.7)

(6.7)

(6.75)

AI

5.6±2.6

6.3±2.5

5.9±2.4

5.8±2.4

5.6±1.7

5.6±1.6

5.7±2.1

5.6±1.9

6.2±3.0

6.3±3.4

5.90±2.6

Correl.

(7.1)

(7.0)

(6.7)

(6.3)

(6.6)

(6.3)

(6.6)

(6.6)

(6.8)

(6.9)

(6.81)

  1. RMSD [Ã…] of ab initio (AI) and template-based (TB) predictions of inter-residue distances as a function of sequence identity to the best template. RMSD is calculated for all residue pairs belonging to the particular protein and then averaged for all proteins in the data set. Values in the brackets are obtained by averaging the obtained RMSDs across the all residue pairs in the dataset.