DeepNGlyPred: A Deep Neural Network-Based Approach for Human N-Linked Glycosylation Site Prediction

Pakhrin, Subash C.; Aoki‐Kinoshita, Kiyoko F.; Caragea, Doina; Kc, Dukka B.

doi:10.3390/molecules26237314

Cited by 17 publications

(16 citation statements)

References 51 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Oligo-saccharyltransferases transfer a preassembled oligosaccharide from a lipid-linked donor to Asn residues within glycosylation acceptor “sequons”: Asn-X-Thr/Ser/Cys, where X is any residue other than Pro. Not all sequons are glycosylated however, reflecting the importance of other primary and secondary structural features (37, 38). Once conjugated, these N-linked oligosaccharides undergo further processing, the products of which are specifically recognized by ER-localized lectins.…”

Section: Resultsmentioning

confidence: 99%

Evolutionary conservation of sequence motifs at sites of protein modification

Dohlman

2022

Preprint

View full text Add to dashboard Cite

Gene duplications are common in biology and are likely to be an important source of functional diversification and specialization. The yeast Saccharomyces cerevisiae underwent a whole genome duplication event early in evolution, and a substantial number of duplicated genes have been retained. We identified more than 3,500 instances where only one of two paralogous proteins undergoes post-translational modification despite having retained the same amino acid residue in both. We also developed a web-based search algorithm (CoSMoS.c.) that scores conservation of amino acid sequences based on 1011 wild and domesticated yeast isolates and used it to compare differentially-modified pairs of paralogous proteins. We found that the most common modifications – phosphorylation, ubiquitylation and acylation but not N-glycosylation – occur in regions of high sequence conservation. Such conservation is evident even for ubiquitylation and succinylation, where there is no established ‘consensus site’ for modification. Differences in phosphorylation were not associated with predicted secondary structure or solvent accessibility, but did mirror known differences in kinase-substrate interactions. By integrating data from large scale proteomics and genomics analysis, in a system with such substantial genetic diversity, we obtained a more comprehensive understanding of the functional basis for genetic redundancies that have persisted for 100 million years.

show abstract

Section: Resultsmentioning

confidence: 99%

Evolutionary conservation of sequence motifs at sites of protein modification

Dohlman

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Nonetheless, the two most recent additions to the available models for the prediction of N -glycosylation have implemented deep learning methods (see Figure A). SPRINT-Gly trained an SVM and deep NNs on calculated amino acid, evolutionary, structural, and physicochemical features, and, in the same vein, DeepNGlyPred is a deep NN, trained on sequence-based features (e.g., gapped dipeptides), predicted structural features, and evolutionary information. Table recapitulates the glycosite predictive tools cited in sections and , as well as in the present one.…”

Section: Next-generation Machine Learningmentioning

confidence: 99%

Glycoinformatics in the Artificial Intelligence Era

Bojar

Lisacek

2022

Chem. Rev.

View full text Add to dashboard Cite

Artificial intelligence (AI) methods have been and are now being increasingly integrated in prediction software implemented in bioinformatics and its glycoscience branch known as glycoinformatics. AI techniques have evolved in the past decades, and their applications in glycoscience are not yet widespread. This limited use is partly explained by the peculiarities of glyco-data that are notoriously hard to produce and analyze. Nonetheless, as time goes, the accumulation of glycomics, glycoproteomics, and glycan-binding data has reached a point where even the most recent deep learning methods can provide predictors with good performance. We discuss the historical development of the application of various AI methods in the broader field of glycoinformatics. A particular focus is placed on shining a light on challenges in glyco-data handling, contextualized by lessons learnt from related disciplines. Ending on the discussion of state-of-the-art deep learning approaches in glycoinformatics, we also envision the future of glycoinformatics, including development that need to occur in order to truly unleash the capabilities of glycoscience in the systems biology era.

show abstract

“…By considering the protein sequences as sentences, Elnaggar et al (ProtTrans) developed a pretrained protein language model (pLM), namely, ProtT5-XL-UniRef50 trained on 2.5 billion protein sequences from the UniRef50 database. The representation learned by this model has been used in various prediction tasks, and the results demonstrate that the information on the evolutionary context of a sequence, contact map, taxonomy, long-range dependencies, protein structure, subcellular localization, physicochemical properties, and function is encoded in their distributed representation. − Thus, a more effective model of phosphorylation site prediction may be established by using the knowledge distilled by this language model.…”

Section: Introductionmentioning

confidence: 99%

LMPhosSite: A Deep Learning-Based Approach for General Protein Phosphorylation Site Prediction Using Embeddings from the Local Window Sequence and Pretrained Protein Language Model

et al. 2023

Self Cite

View full text Add to dashboard Cite

Phosphorylation is one of the most important post-translational modifications and plays a pivotal role in various cellular processes. Although there exist several computational tools to predict phosphorylation sites, existing tools have not yet harnessed the knowledge distilled by pretrained protein language models. Herein, we present a novel deep learning-based approach called LMPhosSite for the general phosphorylation site prediction that integrates embeddings from the local window sequence and the contextualized embedding obtained using global (overall) protein sequence from a pretrained protein language model to improve the prediction performance. Thus, the LMPhosSite consists of two base-models: one for capturing effective local representation and the other for capturing global per-residue contextualized embedding from a pretrained protein language model. The output of these base-models is integrated using a score-level fusion approach. LMPhosSite achieves a precision, recall, Matthew's correlation coefficient, and F1-score of 38.78%, 67.12%, 0.390, and 49.15%, for the combined serine and threonine independent test data set and 34.90%, 62.03%, 0.298, and 44.67%, respectively, for the tyrosine independent test data set, which is better than the compared approaches. These results demonstrate that LMPhosSite is a robust computational tool for the prediction of the general phosphorylation sites in proteins.

show abstract

DeepNGlyPred: A Deep Neural Network-Based Approach for Human N-Linked Glycosylation Site Prediction

Cited by 17 publications

References 51 publications

Evolutionary conservation of sequence motifs at sites of protein modification

Evolutionary conservation of sequence motifs at sites of protein modification

Glycoinformatics in the Artificial Intelligence Era

LMPhosSite: A Deep Learning-Based Approach for General Protein Phosphorylation Site Prediction Using Embeddings from the Local Window Sequence and Pretrained Protein Language Model

Contact Info

Product

Resources

About