HOW MUCH BOVINE RHODOPSIN CRYSTAL STRUCTURE IS

USEFUL FOR MODELING HUMAN GPCRS?

β2-Adrenergic Receptor as a Test Case

Anwar Rayan, Mohamed Hegaze and Jamal Raiyn

QRC-Qasemi Research Center,Al-Qasemi Academic College, P.O.B. 124, Baka El-Garbiah 30100, Israel

Keywords: Data mining of GPCR database, Homology modeling, 3D-structure prediction.

Abstract: Availability of realistic models for human G-Protein Coupled Receptors (hGPCRs) will aid structure-based

drug design (SBDD), thus shortening the time period needed for drug development and minimizing cross-

reactivity of drugs with other hGPCRs. Many researchers have constructed models for hGPCRs with

homology modeling techniques based on the X-ray structure of bovine rhodopsin and recently to β2-

adrenergic receptor which are the only two GPCRs that have high resolution crystal structures. In this study,

we evaluate the usefulness of the bovine rhodopsin crystal structures for modeling hGPCRs by analysis of

large database of human G-protein coupled receptors that are members of family A (rhodopsin family). The

recently released structure of β2-adrenergic receptor was used as a test case for validation purposes of our

findings. From pair-wise sequence alignment of each of the receptors in the database to bovine rhodopsin,

we come to the conclusion that only for few hGPCRs, X-ray structure of rhodopsin could be used as a

template for modeling the trans-membrane domains (TMDs).The detailed analysis of the whole database

shows that in general, similarity to bovine rhodopsin is found more in the middle/endoplasmic part than in

the exoplasmic part. The shift in the cytoplasmic end of TMD-6 that has been seen in the recently released

crystal structure of β2-adrenergic receptor could be understood well from our bioinformatics study. On the

basis of our results from this research, we propose to regard specific parts from the endoplasmic domain of

the rhodopsin helices as appropriate template for constructing models of other GPCRs, while most of the

exoplasmic parts of GPCRs in this family require other techniques for their modeling, due to the low

sequence similarity between the receptors and rhodopsin in that region.

1 INTRODUCTION

G-protein coupled receptors (GPCRs) are membrane

embedded proteins that have a typical structural

topology: seven transmembrane helices (7TMH)

connected by intracellular and extracellular loops,

with an extracellular N-terminal and an intracellular

C-terminal (Gether, 2000). GPCRs derive their name

from their ability to recruit and to regulate the

activity of intracellular heterotrimeric G-proteins.

Their main role is to transfer (transduce) a signal

across the cell membrane. Such signals emerge from

interactions of GPCRs with extracellular agents,

which are highly diverse entities (e.g., ions, biogenic

Abbreviations: hGPCR (human G-Protein Coupled Receptor),

TMD (Trans Membrane Domain).

amines, nucleosides, lipids, peptides, proteins, and

even light). These agents are called “ligands” or

“agonists” and ligand binding is followed by a

change in the state of a GPCR to one with decreased

affinity to G-proteins. Thus, the meeting between

such agonists and GPCRs results in the conversion

of “extracellular events” to intracellular responses

(Nurnberg et al., 1995).

GPCRs are implicated in a very wide range of

body functions and processes, including

cardiovascular, nervous, endocrine, and immune

systems. Also, their involvement in many disease

conditions such as asthma, cardiovascular disease,

central nervous system disorders, pain and others

has been proven or suspected and they are

considered to be the single largest group of drug

targets. It has been estimated that GPCRs comprise

~45% of drug targets (Drews, 2000) and more than

291

Rayan A., Hegaze M. and Raiyn J. (2009).

HOW MUCH BOVINE RHODOPSIN CRYSTAL STRUCTURE IS USEFUL FOR MODELING HUMAN GPCRS? - β2-Adrenergic Receptor as a Test

Case.

In Proceedings of the International Conference on Bio-inspired Systems and Signal Processing, pages 291-298

DOI: 10.5220/0001542402910298

 SciTePress

50% of current drugs are directed to GPCRs (Nambi

and Aiyar, 2003).

The number of known GPCRs is in the

thousands, and many more are being discovered as a

result of recent advances in genomics and

proteomics. To be useful for drug design, structures

of these drug targets should be elucidated, in order

to employ them by methods of “Structure Based

Drug Design” (SBDD). The structural aspects of

GPCRs are however a source of constant debate in

recent years (Bissantz et al., 2003).

Direct experimental study of GPCR structures is

currently too complicated due to their native

membrane environment. Until November 2007,

only a single G-protein-coupled receptor, bovine

rhodopsin, has been studied by high-resolution

crystallography (Palczewski et al., 2000; Okada and

Nakamichi, 2004). β2-adrenergic receptor was the

second GPCR to solve and its structure revealed fair

similarity to the model obtained based on rhodopsin

as a template (Rasmussen et al., 2007; Cherezov et

al., 2007). The prospects for elucidating the

structures of other GPCR are not very high, and

await a major breakthrough. With no other structures

at hand, rhodopsin and/or β2-adrenergic receptor are

considered to be the prototypes of the main family of

GPCRs, of type A.

Due to the lack of experimental 3D-structures of

other GPCRs, one could hope to gain from

approximations based on molecular models. While

“ab initio” modeling is not practical yet for any

protein, “homology” or “comparative” modeling are

quite established methods (Rayan et al., 2000) and

are expected to be especially successful in the GPCR

subfamily A, that is considered to have the general

features of rhodopsin. Indeed, many GPCR

structures have been modelled recently, based on the

template of bovine rhodopsin/β2-adrenergic

receptors, by using its backbone coordinates and

adding the appropriate side chains of each sequence

(Eszter and Zsolt, 2008). Such “homology” or

“comparative” modeling of GPCRs has been aided

mainly by experimental information from point

mutations and other experimental resources

(http://www.gpcr.irg/7tm/, 2006). The length of

helices in the TMD remain similar in the modelled

GPCRs to those of the template rhodopsin, and loops

are not included in the template construction, except

in those rare cases where loop lengths are similar to

those of rhodopsin. But other approaches for

constructing models of GPCRs suggest that GPCRs

could differ in their structure from rhodopsin even

though their general features are similar (Oliveira et

al., 2002).

There are a few indications to justify such

deviations from the rhodopsin structure, in

constructing models for other GPCRs. A review by

Baker and Sali (Schacham et al., 2001) has shown

that a homology model for a protein at medium size

at least and with sequence identity of less than 30%

to the template crystal structure is unreliable. The

averaged sequence identity of bovine rhodopsin to

hGPCRs is less than 20%, meaning a homology-

based approach is unlikely to provide a reliable

structure to be used for making predictions. Others

in the community think that this “rule” is correct in

globular proteins and it is doubtful if this “rule”

could be extended to membrane proteins. Also, this

rule does not specify how identity should be

distributed along a sequence. As much as the GPCRs

super family is united by an overall structural

topology and an ability to recruit and regulate the

activity of G proteins, sequence identity between

super family members, even in the more conserved

transmembrane cores is too low. Significant

sequence conservation is found, however, within

several subfamilies of GPCRs. The subfamily of

rhodopsin-like GPCRs is by far the largest (more

than 85% of GPCRs) and is characterized by the

presence of some 35 (out of ~190) highly conserved

residue positions in the TMD, that may be involved

in binding and/or in activation (Baler and Sali,

2001).

The conserved positions along the TM sequences

constitute less than 20%. In contrast, the intracellular

and extracellular loops and the N- and C- terminals

of GPCRs vary in their lengths and therefore they

pose an alignment problem. Palczewski, K. and his

colleagues (Baldwin et al., 1997) via investigation of

sequence analysis of the TMD of GPCRs

demonstrated that “… the extracellular domain is the

least conserved, while GPCRs display considerable

conservation toward the endoplasmic side...” While

this is an important observation, it lacks specific

quantitative character. The conclusions of that study

concentrated on individual residue conservation and

on microenvironment conservation, and have thus

detected the most conserved residues in the TMD.

The authors concluded by suggesting that “It is

reasonable to speculate that the overall fold of these

receptors is highly conserved”. One of the

implications of that study are thus, that it is

reasonable to use the overall structure of rhodopsin

to model the TMD of other GPCRs.

Therefore, the question remains open, to what

extent is the structure of rhodopsin useful as a

template for constructing models of other GPCRs? A

quantitative measure of conservation in that family

BIOSIGNALS 2009 - International Conference on Bio-inspired Systems and Signal Processing

292

of GPCRs could be helpful for deciding upon the

exact parts of rhodopsin that could be used as

templates for such comparative modeling, and those

that should better be excluded. Should we use the

full extent of TM helices, some of the helices, or

stretches of sequences along helices? It was already

noticed earlier that endoplasmic parts of the TMD

are more conserved than exoplasmic parts (Baldwin

et al., 1997). But what are the quantitative aspects of

that conservation and how do they impinge on the

most important decision, which is - how much of the

rhodopsin structure may be used to model other

GPCRs?

Between the two extreme approaches, to use the

full crystallographic structure of the TMD of

rhodopsin or to employ none of it, we propose an

alternative. From our quantitative analysis, we

assign the parts of the structure of rhodopsin that

may be used as a template, and suggest to construct

the rest by other methods that allow deviations from

the crystal structure of the template.

2 METHODS

In this study, we hope to examine if there is a

quantitative basis for modeling the TMDs of

hGPCRs based on the X-ray structure of bovine

rhodopsin. A database of 951 rhodopsin like

hGPCRs were achieved from RAND

Biotechnologies Ltd company. They have used in-

house software called GPCR-scanner to screen the

protein database of human species composed of

63125 proteins (Ensembl human database). Trans-

membrane domains allocations and multiple

sequence alignments were performed by applying

Intelligent Learning Engine technology (Mirzadegan

et al., 2003) from RAND Biotechnologies Ltd

company. Some of the 951 receptors are identical in

the TMDs and differ only in length of the protein

sequence or in the rest of the structure – the two

terminals and/or the extacellular and/or the

intracellular loops. Higher similarity in sequences

means better chance to have close three-dimensional

structures and high confidence to obtain reliable

model for the query protein.

Sequence Alignments

Stretches of helical sequences for each of the

GPCRs have been determined by TMDs-Scanner

(Rayan and Raiyn, 2008), and were subsequently

aligned with those of rhodopsin in the crystal

structure. The length of each helix was imposed by

the rhodopsin template (TMDs) and is 194 residues

in total. No insertions or deletions were considered.

The calculation of cumulative similarity of

sequences to bovine rhodopsin or any other receptor

is expressed by the average of conservation

scores for single sequences positions:

%100∗=

(1)

Where l is the number of amino acid positions in

the sequence of a helix in the TMD and Cj is the

score of the bovine rhodopsin/or of any other

receptor amino acid at position j and can adopt a

value of 1 if the residues are identical and a value of

0 if the residues are not identical. This score was

calculated in order to evaluate the similarity for the

seven TMDs separately as well as the lower

endoplasmic part (G-protein binding) and the upper

exoplasmic part (ligand binding) of the TMDs or

over certain windows along the helices.

Optimization of Windows’ Positions

After a window width was determined, the first

residue in the helix starts the window and the

identity percentage to rhodopsin was evaluated for a

certain hGPCR. The window was then shifted by

one amino acid all along the helix as well as the

other helices. The evaluation has been performed for

all the hGPCRs in our database. The analysis was

done in a few windows of widths between 7-14

residues. We have concentrated on the results of

windows of 11 residues, which are close to about

three such turns, respectively.

3 RESULTS AND DISCUSSION

Conserved Residues in the TMDs

Looking on the frequencies of individual residues in

particular positions along the TMDs (unpublished

data) reveals that large number of positions are

enriched with certain type of amino acid. Very low

variability in specific position contents could mean

importance in signal transduction pathway or in

structural fold. Those residues are mostly found in

the endoplasmic half of the TMDs or interacting

with the membrane or phospholipids head groups in

the edges of the membrane. The frequencies in some

case are different from those reported by Tara

Mirzadegan et al (Baldwin et al., 1997). For

example, in helix I, Gly20, Leu23 and Val24 were

HOW MUCH BOVINE RHODOPSIN CRYSTAL STRUCTURE IS USEFUL FOR MODELING HUMAN GPCRS? -

ß2-Adrenergic Receptor as a Test Case

293

found 79.2% instead of 68%; 50.2% instead of 60%

and 36.6% instead of 66%, respectively. Position 9

in helix II is occupied by Leu in 92.9% while the

other amino acid types are mostly very hydrophobic

like Ile, Met or Phe. This position could be

important to determine the height of the helix by

fixing this hydrophobic moiety in interaction with

the membrane. Position 16 is occupied in 43.9% by

Ser or Thr which properly interact with Trp from

helix IV. Basic residues are dominant in the first two

positions of helix IV and helix VII. Those residues

and others could play important role in determining

the orientation of the GPCR relative to the

membrane.

Entire Similarity in the TMDs of the hGPCRs

To check the entire similarity between all members

in our database, the receptors were clustered by

requiring that clusters should be dissimilar at least

by x% (with x ranging between 1-100). For

example, assume that the threshold for clustering is

x%, then, if receptor A has sequence identity with

receptor B less than the particular threshold, the two

receptors are considered one cluster.

How much human rhodopsin-like GPCRs are

similar?

200

400

600

800

10 30 50 70 90

Identity percentage threshold

Number of clusters

Figure 1: Each cluster should have at least one pair of

receptors shairing percentage of identity withen the TMD

above a certain threshold. Number of clusters converge to

one near 25% of identity. The horizontal axis shows the

sequence identity threshold while the vertical one shows

the number of clusters.

The process is continued until all pairs of receptors

are evaluated. Each receptor in each cluster should

share sequence identity less than x% with at least

one other receptor. The number of the clusters in

each threshold and the shape of the obtained graph

could be an index for the cumulative sequence

identity within the family or subfamily. If the

number of clusters converge to 1 in high threshold,

then we should conclude that the cumulative

sequence identity is high. Number of clusters

converge to one near 25% of identity in TMDs of

human GPCRs (figure 1) while it is in 42% and

37% of identity in amine and peptide subfamilies

respectively.

Similarity with Bovine Rhodopsin

Firstly, similarities within the TMDs were evaluated,

and then in order to evaluate the similarities in the

upper half (ligand binding domain) and the lower

half (G-protein binding domain) separately, each

helix was divided at its centre. The averaged

similarity of the whole TMDs was 19.7%. While in

the endoplasmic half of the TMDs, it was greater

than in the exoplasmic part. The average score for

the endoplasmic half of the rhodopsin-like hGPCRs

is 25.0% while for the exoplasmic half, it is 14.1%.

Identity percentage to rhodopsin withen

TMDs

101

0 20406080100

Identiry percentage

Receptors count

Figure 2: Pair-wise alignment of each family A receptor in

the human genome with rhodopsin separately (only

TMDs). Only six receptors are above 30% of identity and

according to the well-known rules in the field of

homology modeling, X-ray structure of the TMDs of

rhodopsin could be employed for constructing models

with enough confidence.

From pair-wise alignment of all hGPCRs with

bovine rhodopsin, we obtained only six receptors

with sequence identity above a threshold of 30%.

And as depicted in figure 2, most of the receptors

have sequence identity around 20%. The need for a

detailed analysis of the similarity to bovine

rhodopsin stems from the question of usefulness of

the rhodopsin model as a template for constructing

other GPCRs. Any model construction must relate to

sequential parts of the structure and not to individual

positions in space. Therefore, it is important to

record the change in the similarity along each one of

the helices and to realize which parts may be

BIOSIGNALS 2009 - International Conference on Bio-inspired Systems and Signal Processing

294

considered to be “stable” enough so that a variation

of sequence will not affect their structures. The

conservation of sequence stretches of different

length was calculated. Each stretch begins from N to

In this study, we employed a conservation

scoring of segments in order to examine the extent

of the single known GPCR structure of bovine

rhodopsin which should probably not be “copied” in

modeling of other GPCRs. It was shown previously

that most of the conservation takes place in the

endoplasmic parts of the TMD, but quantitative

evaluations were limited to the conservations of

single residues. In our study, we focused on

cumulative conservation, because structural

templates can not be constructed of isolated residues

that are disconnected. By computing the similarity

along stretches of residues, thus constructing a

“cumulative similarity”, we demonstrated the

quantitative aspects of the differences in

conservation between the more conserved

endoplasmic regions of most TM helices in

rhodopsin-like hGPCRs and the exoplasmic parts.

This has been attributed to the more prominent

structural roles of the endoplasmic parts, or to their

very similar function, to transmit a signal to

intracellular G-proteins. The high variability of the

exoplasmic parts probably reflect the need to interact

and to be specific to a wide range of ligands.

There are certainly other possibilities for

dividing the lengths of the transmembrane helices,

and these may be useful for further refinement. We

have shown that it is possible to determine the exact

number of residues in a “stretch” whose averaged

similarity to bovine rhodopsin does not exceed a

certain threshold. We have also employed the

“windows” method and found that then we could

have better chances to model hGPCRs based on

bovine rhodopsin than employing the whole set of

residues in the endoplasmic half (see figure 3a-3g).

TM-1

0 5 10 15 20 25

First amino acid number

Averaged SI to Rh

Figure 3a

TM-2

0 5 10 15 20 25

First amino acid number

Averaged SI to Rh

Figure 3b

TM-3

0 5 10 15 20 25

First amino acid number

Averaged SI to Rh

Figure 3c

TM-4

0 5 10 15

First amino acid number

Averaged SI to Rh

Figure 3d

TM-5

0 5 10 15 20

First amino acid number

Averaged SI to Rh

Figure 3e

HOW MUCH BOVINE RHODOPSIN CRYSTAL STRUCTURE IS USEFUL FOR MODELING HUMAN GPCRS? -

ß2-Adrenergic Receptor as a Test Case

295

TM-6

0 5 10 15 20 25

First amino acid number

Averaged SI to Rh

Figure 3f

TM-7

0 5 10 15

First amino acid number

Averaged SI to Rh

Figure 3g

Figure 3a-3g: Averaged identity scores to bovine

rhodopsin (equation 1) of the seven TMDs of all family A

hGPCRs averaged over window of 11 residues. Horizontal

axis presents initial window positions. Y-axis is partial

conservation. The direction in each helix goes from the N-

terminal side to the C-terminal side.

Modeling of β2-Adrenergic Receptor based

on Bovine Rhodopsin as a Test Case

Since X-ray structure of β2-adrenergic receptor was

released recently, we have used it to validate our

findings that were obtained in this bioinformatics

study. In figure 4 we find the pairwise sequence

alignment of the transmembranal domains of β2-

adrenergic receptor and Bovine Rodopsin, while in

figure 5, the structural alignment is presented. The

best core segments that were selected according to

the findings as depicted in figure 3 gives backbone

RMSD equal 1.39 Å (see figure 6).

Figure 4: Pair-wise alignment of TMD of bovine

rhodopsin with β2-adrenergic receptor.

Figure 5: Superposition of TMD 3D structures, β2-

adrenergic receptor (2RH1) with Bovine Rodopsin (1F88).

The backbone RMSD is equal 2.05 Å. In general, the

upper half is more deviated than the lower half (mainly the

first three turns of transmembrane-1, left side shown in the

picture above).

BIOSIGNALS 2009 - International Conference on Bio-inspired Systems and Signal Processing

296

Figure 6: Superposition of the 7-transmembranal segments

composed of 11 residues each, that were selected based on

the bioinformatics analysis, β2-adrenergic receptor

(2RH1) with Bovine Rodopsin (1F88). The backbone

RMSD is equal 1.39 Å.

The shift in the cytoplasmic end of TMD-6 that

has been seen in the crystal structure of β2-

adrenergic receptor (Rasmussen et al., 2007) could

be explained by graph 3f. The segment of TMD-6 to

be used for modeling β2-adrenergic receptor based

on bovine rhodopsin in lying on the middle of the

helix.

Pair-wise alignment of the TMDs of family A

hGPCRs with β2-adrenergic receptor is shown in

figure 7. 103 receptors are above 30% of identity

and many others with identity less than 20%. We

will further test if we could obtain better models

while combining segments from the two crystal

structures (bovine rhodopsin and β2-adrenergic

receptor).

Identity Percentage to beta(2) adrenergic within

TMDs

101

121

020406080100

Ide ntity perce ntage

Receptors counts

Figure 7: Pair-wise alignment of each family A receptor in

the human genome with β2-adrenergic receptor separately

(only TMDs). 103 receptors are above 30% of identity.

4 CONCLUSIONS

We present in this study a qualitative and a

quantitative analysis of family A hGPCRs database

and tested the usefulness of employing crystal

structure of bovine rhodopsin as a template for

modeling the TMDs of other receptors from the

same family. In most cases, as shown in figures 3a-

3g, helix terminals display a smaller conservation

than other parts of the helices. This is also found in

most of the endoplasmic, more conserved parts

(except for helix VI that has a larger conservation

value at the middle). These variations could be

connected to the structural changes from helix to

loop at both the endoplasmic and exoplasmic

terminals. Structural analysis of the recently released

structure of β2-adrenergic receptor and superposition

of certain parts from the transmembranal domains

with Bovine Rhodopsin backed our findings that

were obtained by this study.

Since we are using only a partial template from

the TM helical region of bovine rhodopsin or β2-

adrenergic receptor, there still persists an immense

problem of determining the rest of the helical

coordinates. Based on the information extracted

from this study, we plan to use Molecular Dynamics

(MD), Simulated Annealing (SA) or Iterative

Stochastic Elimination (ISE)

(http://www.pdb.org/pdb/explore.do?structureId=1F

88) in order to construct better models for GPCRs,

starting with a partial template of rhodopsin and/or

β2-adrenergic receptor.

ACKNOWLEDGEMENTS

We gratefully acknowledge RAND Biotechnologies

ltd company for providing us with the database of

rhodopsin like hGPCRs and the Trans-membrane

domains allocations module.

REFERENCES

Gether U., (2000). Uncovering molecular mechanisms

involved in activation of G Protein-Coupled

Receptors. Endocr Rev, 21, 90-113

Nurnberg B., Gudermann T., Schultz G., (1995).

Receptors and G proteins as primary components of

transmembrane signal transduction. J Mol Med, 73,

123-132

Drews J., (2000). Drug discovery: a historical perspective.

Science, 287, 1960-1964

HOW MUCH BOVINE RHODOPSIN CRYSTAL STRUCTURE IS USEFUL FOR MODELING HUMAN GPCRS? -

ß2-Adrenergic Receptor as a Test Case

297

Nambi P. & Aiyar N., (2003). G protein-coupled receptors

in drug discovery. Assay and Drug Development

Technologies, 1, 305-310

Bissantz C., Bernard P., Hibert M. & Rognan D., (2003).

Protein-based virtual screening of databases. II. Are

homology modeling of G-protein coupled receptors

suitable targets. Proteins – Structure Function and

Genetics, 50, 5-25

Palczewski K., Kumasaka T., Hori T., Behnke C.A.,

Motoshima H., Fox B.A. Le Trong I., Okada T.,

Stenkamp R.E., Yamamoto M. & Miyano M. (2000).

Crystal structure of rhodopsin: a G-protein coupled

receptor. Science, 289(5480), 739-45

Okada T., Nakamichi H., (2004). X-ray crystallography of

rhodopsin. Phase Transitions, 77, 21-29

Rasmussen S.G.F., Choi H.-J., Rosenbaum D.M., Kobilka

T.S., Thian F.S., Edwards P.C., Burghammer M.,

Ratnala V.R.P., Sanishvili R., Fischetti R.F., Schertler

G.F.X., Weis W.I. & Kobilka B.K. (2007). Crystal

structure of the human β

adrenergic G-protein-

coupled receptor. Nature, 450(7168), 383-387

Cherezov V., Rosenbaum D.M., Hanson M.A., Rasmussen

S.G.F., Thian F.S., Kobilka T.S., Choi H.-J., Kuhn P.,

Weis W.I., Kobilka B.K. & Stevens R.C. (2007).

Science, 23, 318(5854), 1258-1265

Rayan A., Siew N., Cheno Schwartzs S., Matzner Y.,

Bautsch W. Goldblum A., (2000), A novel

computational method for predicting the

transmembranal structure of G-protein coupled

receptors: application to the human C5aR and C3aR.

Receptors Channels, 7(2), 121-137

Eszter H., Zsolt B., (2008), Homology modeling of breast

cancer resistance protein (ABCG2). Journal of

Structural Biology, 162, 63-74

http://www.gpcr.org/7tm/ (June 2006 release (10.0))

Oliveira L., Paiva A.C.M., Vriend G., (2002), Correlated

mutation analyses on very large sequence families.

Chembiochem, 3, 1010-1017

Shacham S., et al. (2001). Modeling the 3D structure of

GPCRs from sequence. Med Res Rev, 21, 472-483

Baker D., Sali A., (2001). Protein structure prediction and

structural genomics. Science 294 (5540), 93-96

Baldwin J.M., Schertler G.F., Unger V.M. (1997). An

alpha-carbon template for the transmembrane helices

in the rhodopsin family of G-protein-coupled

receptors. J Mol Biol, 272, 144-164

Mirzadegan T., Benko G., Filipek S., Palczewski K.,

(2003). Sequence analyses of G-protein-coupled

receptors: Similarities to rhodopsin. Biochemistry, 42,

2759-2767

Rayan A.M., Raiyn J.A., 2008. Intelligent Learning

Engine (ILE) Optimization Technology, provisional

patent

TMDs

-Scanner software

http://www.pdb.org/pdb/explore.do?structureId=1F88

Rayan A., Noy E., Chema D., Levitzki A., Goldblum A.,

2004. Stochastic algorithm for kinase homology model

construction. Current Medicinal Chemistry, 11, 675-

692

BIOSIGNALS 2009 - International Conference on Bio-inspired Systems and Signal Processing

298