Prediction of G-protein Coupled Receptors using Deep Learning: A

Review

Anuj Singh, Arvind Kumar Tiwari

Department of Computer Science & Engineering, Kamla Nehru Institute of Technology, Sultanpur, Uttar Pradesh, India

Keywords: G-protein coupled receptors, glutamate, G proteins, signalling, amino acid composition, guanosine

triphosphate (GTP), guanosine diphosphate (GDP).

Abstract: The biggest super classes of the membrane proteins are G-protein coupled receptors as well as GPCRs are

very significant for drug design goals. GPCRs are sometimes known as heptahelical receptor as well as

seven-transmembrane receptor. GPCRs are accountable for several physicochemical and biological

activities like cellular growth, neurotransmission, smell as well as vision. This paper presents a review

related to current approaches to predict GPCRs. Extensive research on GPCRs have progressed to novel

discoveries that open undiscovered and promising drug design opportunities and efficient drug-targeting G-

protein coupled receptors therapies. This paper concentrates primarily on the process of deep learning to

estimate GPCRs.

1 INTRODUCTION

GPCRs are often addressed to the heptahelical

receptor or even the seven-transmembrane receptor.

A protein found in the cell membrane which also

binds extracellular substances as well as

communicates information to an intracellular

molecule named as a G protein. In cell membranes

of a large variety of species, like animals, crops,

microorganisms as well as invertebrates, GPCRs are

present in cell membranes of a large variety of

species, like animals, crops, microorganisms as well

as invertebrates(Dorsam, 2007). There are various

forms of G Protein-coupled receptor about 1,000

forms are identified by the human genome only.

Robert J. Lefkowitz, an American molecular

biologist, showed the presence of GPCRs in the year

1970s (Strader, 1994).

A wide class of proteins containing as even

transmembrane helical structural motif are GPCRs

(Karnik, 2003). G protein-coupled receptors are

huge number of related proteins which sense

molecules out of cell as well as initiate cellular

reactions. Coupling with G proteins GPCRs mostly

travel throughout the cell membrane seven times.

There seem to be two major pathways of signal

transduction concerning GPCRs one pathway is

cAMP signal and another one pathway is

phosphatidylinositol signal pathway (Gilman, 1987).

When a ligand binds to the GPCRs, it induces

conformational alterations in the G Protein-coupled

receptors that enable this one to serve as a transfer

mechanism for guanine nucleotides. GPCRs are a

big drug priority and about 34% of all licensed drugs

from the FDA target 108 elements of this group

(Hauser, 2018). Another rapidly emerging field of

pharmaceutical science is the long-discovered

relationship among G Protein-coupled receptors and

several endogenous as well as exogenous

compounds (Trzaskowski, 2012).G Protein-coupled

receptors convey extracellular signals through

membrane in plasma of intracellular effectors

through use of heterotrimeric G proteins (Pierce,

2002). GPCRs initiate drastic conformation

alterations disclosing intracellular sites which also

communicate with it as well as activate G proteins.

G proteins correspond to GTPase group which

consist of three subunits, α, β as well as γ, from

which β subunits as well as γ comprise the βγ-

subunits (Cabrera-Vera, 2003). This induces GDP

dissociation bound to a subunit of the Gα as well as

its substitution to the GTP. Gα-GTP along with Gβγ

subunit control downstream effectors (See Figure: 1)

(Harhammer, 1996).

Singh, A. and Tiwari, A.

Prediction of G-protein Coupled Receptors using Deep Learning: A Review.

DOI: 10.5220/0010563200003161

In Proceedings of the 3rd International Conference on Advanced Computing and Software Engineering (ICACSE 2021), pages 101-105

ISBN: 978-989-758-544-9

101

Figure1: G Signal of GPCRs through heterotrimeric G

proteins (Lynch, 2016).

2 RELATED WORKS

In paper (Bhasin, 2004), author used a SVM based

method, GPCRpred, for determination of G-protein

coupled receptor families as well as subfamilies

from dipeptide composition of proteins. When tested

with the help of 5-fold cross-validation, system

categorized GPCRs versus non-GPCRs with 99.5%

accuracy. In addition, the model can estimate 5 main

groups of GPCRs with an aggregate MCC of 0.81 as

well as efficiency of 97.5%. In paper (Liao, 2016),

the authors used physical-chemical properties,

coming from SVM-Prot, reflecting the G-protein

coupled receptors. They used random forest

classification technique to differentiate them from

all other sequences of proteins. In this work the

average classification accuracy was 91.61% as well

as 0.9282 was the average AUC.

G-protein-coupled receptors are main group of

receptors for the cell surface as well as one of the

most active sites of therapeutic drugs. Many of the

G-protein-coupled receptors functions are uncertain

along with identifying their ligands as well as signal

transduction pathways are time intensive and costly.

It prompts everyone meting a crucial problem: how

to develop an integrated method for the

classification of receptors coupled with G-proteins

family to assist in classification of medicines as well

as speed up the medicine research process. It is hard

to estimate the categorisation of G-protein-coupled

receptors by means of traditional sequence

alignment strategies due to their extremely divergent

nature. The covariant discriminant predictor was

implemented to estimate the families of G-protein-

coupled receptors in order to deal with such a

scenario (Chou, 2005). In paper (Rosenbaum, 2009),

the authors illustrated latest structural research of

high-resolution findings have been given into the

molecular pathways of activation as well as

constitutive function of G-protein coupled receptors.

Many of the physiological reactions to hormones,

neurotransmitters and stimulants in the atmosphere

are mediated by GPCRs and as therapeutic goals,

they have tremendous potential for a wide range of

infections. Our sense of hearing, our scent, our taste,

and our discomfort are mediated by GPCRs. They

are active in mechanisms of cell identification as

well as interaction, and have thus developed as an

influential superfamily for drug goals.

Unfortunately, for only one GPCR, the atomic-level

structure is open, making it hard to use structure-

based skills to solve medicines as well as mutation

research. They have currently developed approaches

of first principles for forecasting GPCR structure

and forecasting the sites of ligand binding and

relative binding allegiances (Vaidehi, 2002).

In paper (Basith, 2018), authors concentrated

primarily on the concepts of G-protein coupled

receptors drug development focused on

cheminformatics. They provide a complete analysis

of the cheminformatics methods focused on ligands

and structures, which are best demonstrated by

studies of GPCRs. In addition, an effective fusion of

ligand-based experience and structure-based

experience, like an incorporated solution, is also

addressed, which is appearing as a potential

cheminformatics-based G-protein coupled receptors

drug development technique. The current GPCR

structural biology development offers novel

visibility into ligand binding, conformational

dynamics, as well as signalling results regulation.

With the current techniques to multi-dimensional

drug action development, these insights allow

detailed classification of drugs along with their

pharmacodynamic features, which can be related to

the receptor structure as well as estimates of the

effectiveness of preclinical drugs (Wootten, 2018).

In paper (Popov, 2018), author developed a

comprehensive analytical method called

CompoMug, which uses Sequence-based research,

structural experience as well as a developed model

of machine learning to effectively forecast

stabilizing mutations in GPCRs. In paper (Carpenter,

2016), author illustrated the structure of a minimal G

protein, mini-Gs, which consist only from the

adenylate cyclase triggering G protein Gs of the

GTPase system. Mini-Gs are a thin, soluble protein

that, in absence of sub-units of Gβγ, effectively

combines GPCRs.

Extensive research on GPCRs have linked to

novel developments that open undiscovered and

ICACSE 2021 - International Conference on Advanced Computing and Software Engineering

102

promising drug design prospects and cost effective

drug-targeting GPCR therapies. This included the

development of unique signaling mechanisms like

ligand promiscuity occurring in cross-talks of

multitarget ligands, allosteric modulation and the

development of receptor homo as well as oligomers

that can be analyzed effectively with the help of

analytical modeling. Computer-aided approaches for

drug discovery can lower the price of creating drugs

by approximately 50% (Kaczor, 2016). In paper

(Bartuzi, 2017), the authors concentrated on

advances in docking of G protein-coupled receptors.

Appropriate statistical restoration of real ligand-

receptor method is known as molecular docking. In

paper (Schneider, 2018), the author illustrated latest

discoveries of hybrid coarse-grained membrane

protein approaches. They concentrate on in-house

molecular mechanics/coarse-grained approach. They

demonstrate that our molecular mechanics/coarse-

grained method is capable of capturing the atomistic

information of ligand binding interaction.

In paper (Vaidehi, 2016), authors identified the

existing structure of the analytical approaches that

provide inputs into G protein-coupled receptors

allosteric communication as well as explain how

allosteric modulators can be constructed with this

knowledge. The binding of ligands in the

extracellular region to GPCRs conveys the stimulus

to the intracellular region to activate coupling with

effector proteins. The method of this allosteric

contact appears to be mostly unexplored. In paper

(Foster, 2019), author reported the pairing of

cognate peptides as well as receptors. They define

common features that

reveal additional possible

peptidergic signaling mechanisms by

combining

selective genomics through 313 organisms and

bioinformatics over all protein sequences as well as

architectures of individual class A G protein-coupled

receptors. They fused 17 potential endogenous

ligands with five orphan G protein-coupled receptors

correlated with disorders involving developmental,

nervous as well as reproductive system diseases

employing three orthogonal biochemical assays.

In paper (Kobilka, 2007), the author illustrated

dynamic design of G-protein coupled receptors

activation structure along with functionality mainly

based on spectroscopic analyses of purified human

adrenergic β2 receptor. In paper (Shiraishi, 2019),

author established an original peptide descriptor-

incorporated SVM to estimate 22 pairs of

neuropeptides G protein-coupled receptors. For a 41

p% hit rate, the predicted pair signaling assays

identified 1 homologous neuropeptide and 11 Ciona-

specific neuropeptides G protein-coupled receptors.

In paper (Xiao, 2011), author developed a novel

Predictor by combining pseudo-amino acid

composition functional domainas well as the

pseudo-amino acid composition low-frequency

Fourier range. This novel predictor is named GPCR-

2L, in which "2L" implies a two-layer predictor.

GPCR-2L's total hit rate in recognizing there are

almost 97.2 percent of proteins as GPCRs or non-

GPCRs. In paper (Guo, 2006), author established a

novel technique ACC transform based support

vector machine to determine precision of coupling

among G protein-coupled receptors with G-proteins.

The main sequences of amino acids are converted

into vectors dependent on amino acids' key

physicochemical characteristics as well as the

content is converted by the implementation of ACC

transformation into a uniform matrix. Support vector

machine is qualified and tested through jackknife

testing for nonpromiscuous coupled G protein-

coupled receptors as well as promiscuous coupled G

protein-coupled receptors.

In paper (Chou, 2002), the author developed a

rapid system based on sequences to classify their

various G-protein-coupled receptors. In cellular

signaling networks that control different metabolic

functions like vision, odor, neurotransmission,

inflammation, cellular metabolism and cell

development, G-protein-coupled receptors perform a

major position of significance. For perception of

human physiology as well as sickness, such proteins

are really significant. Several pharmaceutical study

initiatives have tried to explain their composition

and purpose. These are hard to crystallize, and

therefore a few of them seem to not melt in

traditional solvents so very less number of GPCRs

structures have been identified. In paper (Wess,

1997), author outlined current evidence extracted

from structural, molecular genetics, biochemical, as

well as biophysical research that have cast fresh

insight on these processes and seek to combine

them.

3 CONCLUSIONS

The biggest super classes of the membrane proteins

are G-protein coupled receptors as well as GPCRs

are very significant for drug design goals. GPCRs

are sometimes known as heptahelical receptor as

well as seven-transmembrane receptor. GPCRs are

accountable for several physicochemical and

biological activities like cellular growth,

neurotransmission, smell as well as vision. This

paper has presented a review for current approaches

Prediction of G-protein Coupled Receptors using Deep Learning: A Review

103

to predict GPCRs. Extensive research on GPCRs

have progressed to novel discoveries that open

undiscovered and promising drug design

opportunities and efficient drug-targeting G-protein

coupled receptors therapies. This paper has

concentrated primarily on the process of deep

learning to estimate GPCRs.

REFERENCES

Dorsam, R. T., & Gutkind, J. S. (2007). G-protein-coupled

receptors and cancer. Nature reviews cancer, 7(2), 79-

94.

Strader, C. D., Fong, T. M., Tota, M. R., Underwood, D.,

& Dixon, R. A. (1994). Structure and function of G

protein-coupled receptors. Annual review of

biochemistry, 63(1), 101-132.

Karnik, S. S., Gogonea, C., Patil, S., Saad, Y., &

Takezako, T. (2003). Activation of G-protein-coupled

receptors: a common molecular mechanism. Trends in

Endocrinology & Metabolism, 14(9), 431-437.

Gilman, A. G. (1987). G proteins: transducers of receptor-

generated signals. Annual review of biochemistry,

56(1), 615-649.

Hauser, A. S., Chavali, S., Masuho, I., Jahn, L. J.,

Martemyanov, K. A., Gloriam, D. E., & Babu, M. M.

(2018). Pharmacogenomics of GPCR drug targets.

Cell, 172(1-2), 41-54.

Trzaskowski, B., Latek, D., Yuan, S., Ghoshdastider, U.,

Debinski, A., & Filipek, S. (2012). Action of

molecular switches in GPCRs-theoretical and

experimental studies. Current medicinal chemistry,

19(8), 1090-1109.

Pierce, K. L., Premont, R. T., & Lefkowitz, R. J. (2002).

Seven-transmembrane receptors. Nature reviews

Molecular cell biology, 3(9), 639-650.

Cabrera-Vera, T. M., Vanhauwe, J., Thomas, T. O.,

Medkova, M., Preininger, A., Mazzoni, M. R., &

Hamm, H. E. (2003). Insights into G protein structure,

function, and regulation. Endocrine reviews, 24(6),

765-781.

Harhammer, R., Gohla, A., & Schultz, G. (1996).

Interaction of G protein Gβγ dimers with small

GTP‐binding proteins of the Rho family. FEBS letters,

399(3), 211-214.

Lynch, J. R., & Wang, J. Y. (2016). G protein-coupled

receptor signaling in stem cells and cancer.

International journal of molecular sciences, 17(5), 707.

Bhasin, M., & Raghava, G. P. S. (2004). GPCRpred: an

SVM-based method for prediction of families and

subfamilies of G-protein coupled receptors. Nucleic

acids research, 32(suppl_2), W383-W389.

Liao, Z., Ju, Y., & Zou, Q. (2016). Prediction of G

protein-coupled receptors with SVM-prot features and

random forest. Scientifica, 2016.

Chou, K. C. (2005). Prediction of G-protein-coupled

receptor classes. Journal of proteome research, 4(4),

1413-1418.

Rosenbaum, D. M., Rasmussen, S. G., & Kobilka, B. K.

(2009). The structure and function of G-protein-

coupled receptors. Nature, 459(7245), 356-363.

Vaidehi, N., Floriano, W. B., Trabanino, R., Hall, S. E.,

Freddolino, P., Choi, E. J., ... & Goddard, W. A.

(2002). Prediction of structure and function of G

protein-coupled receptors. Proceedings of the National

Academy of Sciences, 99(20), 12622-12627.

Basith, S., Cui, M., Macalino, S. J., Park, J., Clavio, N. A.,

Kang, S., & Choi, S. (2018). Exploring G protein-

coupled receptors (GPCRs) ligand space via

cheminformatics approaches: impact on rational drug

design. Frontiers in pharmacology, 9, 128.

Wootten, D., Christopoulos, A., Marti-Solano, M., Babu,

M. M., & Sexton, P. M. (2018). Mechanisms of

signalling and biased agonism in G protein-coupled

receptors. Nature reviews Molecular cell biology,

19(10), 638-653.

Popov, P., Peng, Y., Shen, L., Stevens, R. C., Cherezov,

V., Liu, Z. J., & Katritch, V. (2018). Computational

design of thermostabilizing point mutations for G

protein-coupled receptors. Elife, 7, e34729.

Carpenter, B., & Tate, C. G. (2016). Engineering a

minimal G protein to facilitate crystallisation of G

protein-coupled receptors in their active conformation.

Protein Engineering, Design and Selection, 29(12),

583-594.

Kaczor, A. A., Rutkowska, E., Bartuzi, D., Targowska-

Duda, K. M., Matosiuk, D., & Selent, J. (2016).

Computational methods for studying G protein-

coupled receptors (GPCRs). Methods in cell biology,

132, 359-399.

Bartuzi, D., Kaczor, A. A., Targowska-Duda, K. M., &

Matosiuk, D. (2017). Recent advances and

applications of molecular docking to G protein-

coupled receptors. Molecules, 22(2), 340.

Schneider, J., Korshunova, K., Musiani, F., Alfonso-

Prieto, M., Giorgetti, A., & Carloni, P. (2018).

Predicting ligand binding poses for low-resolution

membrane protein models: Perspectives from

multiscale simulations. Biochemical and biophysical

research communications, 498(2), 366-374.

Vaidehi, N., & Bhattacharya, S. (2016). Allosteric

communication pipelines in G-protein-coupled

receptors. Current opinion in pharmacology, 30, 76-

83.

Foster, S. R., Hauser, A. S., Vedel, L., Strachan, R. T.,

Huang, X. P., Gavin, A. C., ... & Gloriam, D. E.

(2019). Discovery of human signaling systems: pairing

peptides to G protein-coupled receptors. Cell, 179(4),

895-908.

Kobilka, B. K., & Deupi, X. (2007). Conformational

complexity of G-protein-coupled receptors. Trends in

pharmacological sciences, 28(8), 397-406.

Shiraishi, A., Okuda, T., Miyasaka, N., Osugi, T., Okuno,

Y., Inoue, J., & Satake, H. (2019). Repertoires of G

protein-coupled receptors for Ciona-specific

neuropeptides. Proceedings of the National Academy

of Sciences, 116(16), 7847-7856.

ICACSE 2021 - International Conference on Advanced Computing and Software Engineering

104

Xiao, X., Wang, P., & Chou, K. C. (2011). GPCR-2L:

predicting G protein-coupled receptors and their types

by hybridizing two different modes of pseudo amino

acid compositions. Molecular Biosystems, 7(3), 911-

919.

Guo, Y., Li, M., Lu, M., Wen, Z., & Huang, Z. (2006).

Predicting G‐protein coupled receptors–G‐protein

coupling specificity based on autocross‐covariance

transform. Proteins: structure, function, and

bioinformatics, 65(1), 55-60.

Chou, K. C., & Elrod, D. W. (2002). Bioinformatical

analysis of G-protein-coupled receptors. Journal of

proteome research, 1(5), 429-433.

Wess, J. (1997). G‐protein‐coupled receptors: molecular

mechanisms involved in receptor activation and

selectivity of G‐protein recognition. The FASEB

Journal, 11(5), 346-354.

Prediction of G-protein Coupled Receptors using Deep Learning: A Review

105