Skip to main content

Table 3 Predictive power of features on RB144 data set.

From: Exploiting structural and topological information to improve prediction of RNA-protein binding sites

AUC

Type

Size

Feature

0.65

-

1

Retention coefficient (RC)

0.66

sequential

11

Averaged retention coefficient (saRC)

0.69

topological

19

Averaged retention coefficient (taRC)

0.69

spatial

19

Averaged retention coefficient (aRC)

0.70

-

1

Accessible surface area (ASA)

0.69

spatial

19

Averaged Accessible surface area (aASA)

0.69

-

1

Relative accessible surface area (rASA)

0.68

spatial

19

Averaged Relative accessible surface area (arASA)

0.66

spatial

19

Density (D)

0.56

-

1

Betweenness centrality (BC)

0.71

topological

19

Averaged betweenness centrality (aBC)

0.69

-

1

Status (S)

0.69

topological

19

Averaged Status (aS)

0.62

-

1

Cluster coefficient (CC)

0.64

topological

19

Averaged cluster coefficient (aCC)

0.64

-

1

Degree (G)

0.64

topological

19

Averaged degree (aG)

0.63

-

1

Eccentricity (E)

0.64

topological

19

Averaged eccentricity (aE)

0.62

-

1

Closeness (C)

0.64

topological

19

Averaged closeness (aC)

  1. Predictive power (AUC) of features on RB144 data set. Patch type and patch size are listed in columns two and three. A missing patch type and patch size of one indicate features evaluated for individual residues (no patch used).