BackgroundThe characterization of proteins in families and subfamilies, at different levels, entails the definition and use of class labels. When the adscription of a protein to a family is uncertain, or even wrong, this becomes an instance of what has come to be known as a label noise problem. Label noise has a potentially negative effect on any quantitative analysis of proteins that depends on label information. This study investigates class C of G protein-coupled receptors, which are cell membrane proteins of relevance both to biology in general and pharmacology in particular. Their supervised classification into different known subtypes, based on primary sequence data, is hampered by label noise. The latter may stem from a combination of expert knowledge limitations and the lack of a clear correspondence between labels that mostly reflect GPCR functionality and the different representations of the protein primary sequences.ResultsIn this study, we describe a systematic approach, using Support Vector Machine classifiers, to the analysis of G protein-coupled receptor misclassifications. As a proof of concept, this approach is used to assist the discovery of labeling quality problems in a curated, publicly accessible database of this type of proteins. We also investigate the extent to which physico-chemical transformations of the protein sequences reflect G protein-coupled receptor subtype labeling. The candidate mislabeled cases detected with this approach are externally validated with phylogenetic trees and against further trusted sources such as the National Center for Biotechnology Information, Universal Protein Resource, European Bioinformatics Institute and Ensembl Genome Browser information repositories.ConclusionsIn quantitative classification problems, class labels are often by default assumed to be correct. Label noise, though, is bound to be a pervasive problem in bioinformatics, where labels may be obtained indirectly through complex, many-step similarity modelling processes. In the case of G protein-coupled receptors, methods capable of singling out and characterizing those sequences with consistent misclassification behaviour are required to minimize this problem. A systematic, Support Vector Machine-based method has been proposed in this study for such purpose. The proposed method enables a filtering approach to the label noise problem and might become a support tool for database curators in proteomics.Electronic supplementary materialThe online version of this article (doi:10.1186/s12859-015-0731-9) contains supplementary material, which is available to authorized users.

No abstract

No abstract

This brief paper addresses the problem of visually assessing the natural discriminability of the different subtypes that characterize class C G-Protein-Coupled Receptors, which are membrane proteins of interest in pharmacology, from diverse transformations of their primary amino acid sequences. Visualization, achieved using nonlinear dimensionality reduction techniques, is used as an exploratory tool to increase the interpretability of the data subtype structure. Combined with a quantitative estimation of subtype overlapping, this approach should be of help to experts in proteomics.Peer ReviewedPostprint (published version

G-protein-coupled receptors are cell membrane proteins of great interest in biology and pharmacology. Previous analysis of Class C of these receptors has revealed the existence of an upper boundary on the accuracy that can be achieved in the classification of their standard subtypes from the unaligned transformation of their primary sequences. To further investigate this apparent boundary, the focus of the analysis in this paper is placed on receptor sequences that were previously misclassified using supervised learning methods. In our experiments, these sequences are visualized using a nonlinear dimensionality reduction technique and phylogenetic trees. They are subsequently characterized against the rest of the data and, particularly, against the rest of cases of their own subtype. This exploratory visualization should help us to discriminate between different types of misclassification and to build hypotheses about database quality problems and the extent to which GPCR sequence transformations limit subtype discriminability. The reported experiments provide a proof of concept for the proposed method.