Skip to main content

Table 1 Structural descriptors and associated values or thresholds used by EvDTree to partition the root cluster c0 for each residue type into two subclusters (see Figure 2). The EvDTreeDS columns refer to the decision tree classification of a learning set containing only small disulfide-rich proteins (see text).

From: EvDTree: structure-dependent substitution profiles based on decision tree classification of 3D environments

Amino acida

EvDTree

EvDTreeDS

 

Structural descriptor

Value

Structural descriptor

Value

  

subcluster 1

subcluster 2

 

subcluster 1

subcluster 2

Cb

pac

<= 8

> 8

pac

= 0

> 0

Ab

pac

<= 10

> 10

pac

<= 35

> 35

I

pol

<= 41

> 41

C

<= 35

> 35

Lc

pol

<= 53

> 53

ss1

= 6

≠ 6

Fc

C

<= 65

> 65

ss2

= 2

≠ 2

Yc

C

<= 65

> 65

ss2

= 3

≠ 3

Hc

C

<= 35

> 35

ss2

= 3

≠ 3

V

pol

<= 46

> 46

   

M

   

ss1

= 4

≠ 4

W

bur

<= 155

> 155

N

<= 1

> 1

Nd

ss2

= 1

≠ 1

pac

<= 49

> 49

Dd

ss2

= 1

≠ 1

bur

<= 25

> 25

Ed

ss1

= 4

≠ 4

pac

<= 54

> 54

Kd

ss2

= 2

≠ 2

pac

<= 53

> 53

Rd

ss1

= 1

≠ 1

bur

<= 101

> 101

S

pac

<= 3

> 3

   

T

   

pac

<= 49

> 49

G

   

ss1

= 7

≠ 7

Q

   

ss2

= 3

≠ 3

P

   

C

<= 7

> 7

  1. aOnly dichotomies for which the structural information improves the Mean Rank evaluation are shown (Tables 2 and 3). bResidues for which the same descriptor is used in EvDTree and EvDTreeDS. cResidues for which a descriptor corresponding to the local structure has been used by the EvDTreeDS but not by the EvDTree learning process. d Residues for which a descriptor corresponding to the local structure has been used by the EvDTree but not by the EvDTreeDS learning process.
Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us