BIO-GATS: A Tool for Automated GPCR Template Selection Through a Biophysical Approach for Homology Modeling

Abstract

G protein-coupled receptors (GPCRs) are the largest family of membrane proteins with more than 800 members. GPCRs are involved in numerous physiological functions within the human body and are the target of more than 30% of the United States Food and Drug Administration (FDA) approved drugs. At present, over 400 experimental GPCR structures are available in the Protein Data Bank (PDB) representing 76 unique receptors. The absence of an experimental structure for the majority of GPCRs demand homology models for structure-based drug discovery workflows. The generation of good homology models requires appropriate templates. The commonly used methods for template selection are based on sequence identity. However, there exists low sequence identity among the GPCRs. Sequences with similar patterns of hydrophobic residues are often structural homologs, even with low sequence identity. Extending this, we propose a biophysical approach for template selection based principally on hydrophobicity correspondence between the target and the template. Our approach takes into consideration other relevant parameters, including resolution, similarity within the orthosteric binding pocket of GPCRs, and structure completeness, for template selection. The proposed method was implemented in the form of a free tool called Bio-GATS, to provide the user with easy selection of the appropriate template for a query GPCR sequence. Bio-GATS was successfully validated with recent published benchmarking datasets. An application to an olfactory receptor to select an appropriate template has also been provided as a case study.

Machine Learning Assisted Approach for Finding Novel High Activity Agonists of Human Ectopic Olfactory Receptors

“…To identify the top candidates for experimental validation, we conducted induced-fit docking of the highly probable compounds based on 3D homology models of OR1A1 and OR2W1. We built the homology models of OR2W1 (UniProtID: Q9Y3N9) and OR1A1 (UniProtID: Q9P1Q5) using the approach described previously [ 50 ], with the X-ray crystallographic structure of bovine rhodopsin (PDB ID: 1U19) [ 51 ] as a template, and Bio-GATS TM alignments [ 30 ]. The experimental structure of an insect OR has recently been resolved at 3.3 Å resolution [ 9 ] with an inverted topology to human ORs.…”

“…Briefly, the sequences of OR1A1 and OR2W1 were aligned with bovine rhodopsin, based on conserved GPCR motifs ( Supplementary Figure S1 ). The predicted transmembrane domains in both receptors were based on the GRoSS sequence alignment of all known GPCRs sequences [ 52 ], as implemented by Bio-GATS [ 30 ]. Homology modelling was performed using MODELLER 9.18 [ 53 ].…”

“…However, the AlphaFold models generated in this study need a lot of adjustment in the orientation of the transmembrane helices (Jabeen and Ranganathan, unpublished data), to recover the OR binding sites that were validated by published mutational studies [ 20 , 21 , 22 , 23 , 24 ]. Instead, we recently showed that using a biophysical approach, Bio-GATS, for template selection generates an excellent homology model for OR1A1 [ 30 ].…”

“…We selected three methods, RF, SVM and NB, based on their performance for OR1G1 [ 28 ]. We filtered the predicted compounds using knowledge-based homology models of the two receptors, based on the best Bio-GATS template [ 30 ]. From the shortlisted predictions, we validated some of the randomly selected predicted compounds using an in vitro functional assay.…”

MUG: A mutation overview of GPCR subfamily A17 receptors

“…In our study, we developed a database, MUG (mutations understanding GPCRs), which provides an overview of already described mutations and their effect on receptors of subfamily A17. Similar to all class A GPCRs, the members of subfamily A17 share a common architecture of seven transmembrane helices (7 TMs) connected through three extracellular (ECL1–3) and three intracellular loops (ICL1–3) as well as an extracellular N-terminus and an intracellular C-terminus [21] , [22] . The subfamily A17 comprises receptors that bind to biogenic amines [23] , including dopamine receptors (D 1–5 R), serotonin receptors (5-HT 2A-C R, 5-HT 6 R), trace amine receptors (TA 1–3 R, TA 5–6 R, TA 8–9 R), and adrenergic receptors (α 1A/1B/1D -adrenoceptor, α 2A/2B/2 C -adrenoceptor, β 1/2/3 -adrenoceptor).…”