O-GlcNAcylation Prediction: An Unattained Objective

Abstract

Background: O-GlcNAcylation is an essential post-translational modification (PTM) in mammalian cells. It consists in the addition of a N-acetylglucosamine (GlcNAc) residue onto serines or threonines by an O-GlcNAc transferase (OGT). Inhibition of OGT is lethal, and misregulation of this PTM can lead to diverse pathologies including diabetes, Alzheimer's disease and cancers. Knowing the location of O-GlcNAcylation sites and the ability to accurately predict them is therefore of prime importance to a better understanding of this process and its related pathologies. Purpose: Here, we present an evaluation of the current predictors of O-GlcNAcylation sites based on a newly built dataset and an investigation to improve predictions. Methods: Several datasets of experimentally proven O-GlcNAcylated sites were combined, and the resulting meta-dataset was used to evaluate three prediction tools. We further defined a set of new features following the analysis of the primary to tertiary structures of experimentally proven O-GlcNAcylated sites in order to improve predictions by the use of different types of machine learning techniques. Results: Our results show the failure of currently available algorithms to predict O-GlcNAcylated sites with a precision exceeding 9%. Our efforts to improve the precision with new features using machine learning techniques do succeed for equal proportions of O-GlcNAcylated and non-O-GlcNAcylated sites but fail like the other tools for real-life proportions where ~1.4% of S/T are O-GlcNAcylated. Conclusion: Present-day algorithms for O-GlcNAcylation prediction narrowly outperform random prediction. The inclusion of additional features, in combination with machine learning algorithms, does not enhance these predictions, emphasizing a pressing need for further development. We hypothesize that the improvement of prediction algorithms requires characterization of OGT's partners.

Exploration of O-GlcNAc-transferase (OGT) glycosylation sites reveals a target sequence compositional bias

“…4b). In contrast, comparing the number of sites predicted by O-GlcNAcPredII, considered to be the best existing predictor(45), gave a Pearson correlation of 0.59 with a p value of 0.21 for this set. The data for individual mutants showed a distribution in the number of glycosylation sites, matching expectation.…”

“…Nevertheless, a scan of O-GlcNAcylated peptide sequences in the PhosphoSite database (38) indicates that most substrates fall outside of this definition, as well as definitions put forward by other groups. Computational methods that make use of machine learning or neural networks to predict sites of O-GlcNAc modification (23,(39)(40)(41)(42)(43)(44)(45)(46) have been used to address this shortcoming. These computational methods take both sequence and amino acid combinations into account when making their predicitions.…”

Exploration of O-GlcNAc transferase glycosylation sites reveals a target sequence compositional bias

“…Thus, when evaluating O-GlcNAc site predictors, sensitivity is a critical parameter. Taking sensitivity into consideration, O-GlcNAcPRED-II seems to outperform other prediction methods ( 41 , 45 ). Despite the extensive effort put into these computational methods, they yield many false positive and false negative sites, so experimental validation is still necessary ( 47 , 48 ).…”

“…Nevertheless, a scan of O-GlcNAcylated peptide sequences in the PhosphoSite database ( 38 ) indicates that most substrates fall outside of this definition, as well as definitions put forward by other groups. Computational methods that make use of machine learning or neural networks to predict sites of O-GlcNAc modification ( 23 , 39 , 40 , 41 , 42 , 43 , 44 , 45 , 46 ) have been used to address this shortcoming. These computational methods take both sequence and amino acid combinations into account when making their predictions.…”

“…Particular attention must be paid to sensitivity when evaluating these O-GlcNAc predictors, because only a small minority of serines and threonines are expected to be modified. While many proteins can be modified by OGT, the proportion of individual serines and threonines that are modified may be as low as 1.4% ( 45 ). The small number of modified residues makes it easy to achieve a high level of accuracy (combined proportion of correctly identified positive and negative sites) by setting a very high threshold for positive-site identification.…”