Classification of G‐protein coupled receptors by alignment‐independent extraction of principal chemical properties of primary amino acid sequences

Abstract

We have developed an alignment-independent method for classification of G-protein coupled receptors (GPCRs) according to the principal chemical properties of their amino acid sequences. The method relies on a multivariate approach where the primary amino acid sequences are translated into vectors based on the principal physicochemical properties of the amino acids and transformation of the data into a uniform matrix by applying a modified autocross-covariance transform. The application of principal component analysis to a data set of 929 class A GPCRs showed a clear separation of the major classes of GPCRs. The application of partial least squares projection to latent structures created a highly valid model (cross-validated correlation coefficient, Q 2 ‫ס‬ 0.895) that gave unambiguous classification of the GPCRs in the training set according to their ligand binding class. The model was further validated by external prediction of 535 novel GPCRs not included in the training set. Of the latter, only 14 sequences, confined in rapidly expanding GPCR classes, were mispredicted. Moreover, 90 orphan GPCRs out of 165 were tentatively identified to GPCR ligand binding class. The alignment-independent method could be used to assess the importance of the principal chemical properties of every single amino acid in the protein sequences for their contributions in explaining GPCR family membership. It was then revealed that all amino acids in the unaligned sequences contributed to the classifications, albeit to varying extent; the most important amino acids being those that could also be determined to be conserved by using traditional alignment-based methods.The functional and structural annotation of proteins is an important task in proteomics (O'Donovan et al. 2001). There is a strong need for efficient and reliable methods for the analysis of protein sequence data. Existing methods rely mainly on alignment-and similarity-based comparisons. Basing the analysis on common patterns and profiles may implicitly take into account that the structure and function of proteins are determined by the physicochemical properties of their sequence constituents. However, a method that uses a direct quantitative measure of the physicochemical properties of the amino acids would seem a more rational approach. For example, proteins created by divergent or convergent evolution may lack obvious sequence similarity, although they share similar structural organization and biological properties. In such a situation, attempts to align protein sequences might produce ambiguous results or fail.In the past, quantitative descriptions of peptide sequences have been attempted using physicochemical z-scales (Hell-

An Automated System for the Analysis of G Protein-Coupled Receptor Transmembrane Binding Pockets: Alignment, Receptor-Based Pharmacophores, and Their Application

“…Huang 64 recently published that the TM3 helix position (3.32) and the TM7 helix position (7.40) are sufficient for classifying the aminergic GPCRs and for predicting novel members of this class among orphan GPCR sequences. Prusis et al [66][67][68][69] reported a proteochemometrics approach to identify amino acid motifs important for ligand binding to aminergic GPCR receptors from within the TM regions. They described the aligned 7TM helices of the aminergic GPCRs by physicochemical descriptors and applied a multivariate analysis to explore the interactions of the TM helix amino acids of the R 1a -, R 1b -, and R 1d -adrenoreceptors with 4-piperidyl oxazole antagonists.…”

The influence of alignment-free sequence representations on the semi-supervised classification of class C G protein-coupled receptors

“…Among these, some rely on transformations based on the amino acid physicochemical characteristics, such as the ACC transformation [22,41] and the mean transformation [26]. Some advantages of these representations are that: a) they can help to discover divergent receptor genes [20]; b) they do not require a homologous relationship among similar sequences to be assumed; and c) their transformed output can directly be used by standard pattern recognition and machine learning methods.…”

“…In the current study, we use several of them to transform the data sequences, including the Auto-Cross Covariance (ACC) transform [22] and the physicochemical distance transformation (PDT: [23]). In a less complex procedure described in [10] and [26], each GPCR is represented using the means of five z-scale physicochemical descriptors over the sequence.…”

Ligand Modulation of the Epstein-Barr Virus-induced Seven-transmembrane Receptor EBI2

“…Given that this receptor does not have a close homolog, the chemical nature of such a ligand is difficult to infer. Accordingly, EBI2 has been placed in varying 7TM receptor subgroups by different phylogenetic analyses as being a target of peptide (1,5) or lipid ligands (6). Nevertheless, many aspects of EBI2 signaling have been revealed.…”