Prediction of O-glycosylation sites based on multi-scale composition of amino acids and feature selection

Prediction of human O-linked glycosylation sites using stacked generalization and embeddings from pre-trained protein language model

MOTIVATION

O-linked glycosylation, an essential post-translational modification process in Homo sapiens, involves attaching sugar moieties to the oxygen atoms of serine and/or threonine residues. It influences various biological and cellular functions. While threonine or serine residues within protein sequences are potential sites for O-linked glycosylation, not all serine and/or threonine residues undergo this modification, underscoring the importance of characterizing its occurrence. This study presents a novel approach for predicting intracellular and extracellular O-linked glycosylation events on proteins, which are crucial for comprehending cellular processes. Two base multi-layer perceptron models were trained by leveraging a stacked generalization framework. These base models respectively use ProtT5 and Ankh O-linked glycosylation site-specific embeddings whose combined predictions are used to train the meta-multi-layer perceptron model. Trained on extensive O-linked glycosylation datasets, the stacked-generalization model demonstrated high predictive performance on independent test datasets. Furthermore, the study emphasizes the distinction between nucleocytoplasmic and extracellular O-linked glycosylation, offering insights into their functional implications that were overlooked in previous studies. By integrating the protein language model's embedding with stacked generalization techniques, this approach enhances predictive accuracy of O-linked glycosylation events and illuminates the intricate roles of O-linked glycosylation in proteomics, potentially accelerating the discovery of novel glycosylation sites.

RESULTS

Stack-OglyPred-PLM produces Sensitivity, Specificity, Matthews Correlation Coefficient, and Accuracy of 90.50%, 89.60%, 0.464, and 89.70%, respectively on a benchmark NetOGlyc-4.0 independent test dataset. These results demonstrate that Stack-OglyPred-PLM is a robust computational tool to predict O-linked glycosylation sites in proteins.

AVAILABILITY AND IMPLEMENTATION

The developed tool, programs, training, and test dataset are available at https://github.com/PakhrinLab/Stack-OglyPred-PLM.

O-glycosylation site prediction for Homo sapiens by combining properties and sequence features with support vector machine

O-glycosylation is a protein posttranslational modification important in regulating almost all cells. It is related to a large number of physiological and pathological phenomena. Recognizing O-glycosylation sites is the key to further investigating the molecular mechanism of protein posttranslational modification. This study aimed to collect a reliable dataset on Homo sapiens and develop an O-glycosylation predictor for Homo sapiens, named Captor, through multiple features. A random undersampling method and a synthetic minority oversampling technique were employed to deal with imbalanced data. In addition, the Kruskal-Wallis (K-W) test was adopted to optimize feature vectors and improve the performance of the model. A support vector machine, due to its optimal performance, was used to train and optimize the final prediction model after a comprehensive comparison of various classifiers in traditional machine learning methods and deep learning. On the independent test set, Captor outperformed the existing O-glycosylation tool, suggesting that Captor could provide more instructive guidance for further experimental research on O-glycosylation. The source code and datasets are available at https://github.com/YanZhu06/Captor/.

Ridge regression estimated linear probability model predictions of O-glycosylation in proteins with structural and sequence data

BACKGROUND

To-date, no claim regarding finding a consensus sequon for O-glycosylation has been made. Thus, predicting the likelihood of O-glycosylation with sequence and structural information using classical regression analysis is quite difficult. In particular, if a binary response is used to distinguish between O-glycosylated and non-O-glycosylated sequences, an appropriate set of non-O-glycosylatable sequences is hard to find.

RESULTS

Three sequences from similar post-translational modifications (PTMs) of proteins occurring at, or very near, the S/T-site are analyzed: N-glycosylation, O-mucin type (O-GalNAc) glycosylation, and phosphorylation. Results found include: 1) The consensus composite sequon for O-glycosylation is: ~(W-S/T-W), where "~" denotes the "not" operator. 2) The consensus sequon for phosphorylation is ~(W-S/T/Y/H-W); although W-S/T/Y/H-W is not an absolute inhibitor of phosphorylation. 3) For linear probability model (LPM) estimation, N-glycosylated sequences are good approximations to non-O-glycosylatable sequences; although N - ~P - S/T is not an absolute inhibitor of O-glycosylation. 4) The selective positioning of an amino acid along the sequence, differentiates the PTMs of proteins. 5) Some N-glycosylated sequences are also phosphorylated at the S/T-site in the N - ~P - S/T sequon. 6) ASA values for N-glycosylated sequences are stochastically larger than those for O-GlcNAc glycosylated sequences. 7) Structural attributes (beta turn II, II´, helix, beta bridges, beta hairpin, and the phi angle) are significant LPM predictors of O-GlcNAc glycosylation. The LPM with sequence and structural data as explanatory variables yields a Kolmogorov-Smirnov (KS) statistic of 99%. 8) With only sequence data, the KS statistic erodes to 80%, and 21% of out-of-sample O-GlcNAc glycosylated sequences are mispredicted as not being glycosylated. The 95% confidence interval around this mispredictions rate is 16% to 26%.

CONCLUSIONS

The data indicates the existence of a consensus sequon for O-glycosylation; and underscores the germaneness of structural information for predicting the likelihood of O-glycosylation.