Skip to main content

Table 4 Fingerprint discovery results. Coverage values of profile Hidden Markov models (pHMMs) given after each alignment of entire loop sequence regions in comparison to maximized Coverage values of highly discriminative pHMMs derived from selected sub-alignments within low-entropy regions. pHMMs derived from sub-alignments showed up to 12-fold increase in their discriminative power, as measured by Coverage, in comparison to pHMMs that characterize the entire loop sequence alignments. Alternative alignment regions with the same discriminative power are separated by commas, double dots separate beginning and ending of sub-alignments used to generate HMMs. In the case of discovery of non-overlapping sequence fragments with high discriminative power (e.g. Gs loop1), separate pHMMs were generated and appended in the refined library of that group.

From: A method for the prediction of GPCRs coupling specificity to G-proteins using refined profile Hidden Markov Models

Coupling preference

Gi/o

Gq/11

Gs

Intracellular regions

loop1

loop1

loop1

Whole sequence region

1..39/2.65

1..28/16.05

1..59/58.01

Non-overlapping most discriminative regions

3..24/19.86

13..25/30.53

7..19/63.35,43..51/48.09

 

loop2

loop2

loop2

 

1..50/19.20

1..42/0.70

1..39/12.21

 

10..18/32.78

14..34/19.85

18..25/54.19

 

loop3a

loop3a

loop3a

 

1..29/19.54

1..17/16.79

1..15/27.48

 

2..18/27.15

5..17/19.08

6..15/42.74

 

loop3b

loop3b

loop3b

 

1..26/6.63

1..21/3.82

1..15/20.61

 

14..21/23.51

8..19,9..20/12.21

1..15/20.61

 

loop3c

loop3c

loop3c

 

1..29/8.28

1..17/3.81

1..22/0.76

 

8..21/36.09

2..12/48.85

8..21,13..21/10.68

 

C-terminal

C-terminal

C-terminal

 

1..29/8.28

1..24/12.97

1..29/29.77

 

8..21/36.09

5..19/35.87

8..16/57.25