Deep learning based prediction of species-specific protein S-glutathionylation sites

Li, Shihua; Yu, Kai; Wang, Dawei; Liu, Zexian; Zhao, Lifeng; Cheng, Han

doi:10.1016/j.bbapap.2020.140422

Cited by 22 publications

(45 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition, the results of ablation experiments based on feature combinations and model architecture indicated that using multilane dense convolutional attention network to extract sequence information, physicochemical properties of amino acids, and structural properties of proteins can help to transform the original sequence fragments into meaningful abstract representations, thereby further helping the model to better complete the lysine succinylation prediction. In the future, we will try to adopt more feature representations (such as a position-specific scoring matrix [16,[52][53][54] or protein function features [55]) and explore other deep learning networks (such as a capsule network [56,57] or improved CNN models [58]) for the prediction of succinylation sites.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

MDCAN-Lys: A Model for Predicting Succinylation Sites Based on Multilane Dense Convolutional Attention Network

et al. 2021

View full text Add to dashboard Cite

Lysine succinylation is an important post-translational modification, whose abnormalities are closely related to the occurrence and development of many diseases. Therefore, exploring effective methods to identify succinylation sites is helpful for disease treatment and research of related drugs. However, most existing computational methods for the prediction of succinylation sites are still based on machine learning. With the increasing volume of data and complexity of feature representations, it is necessary to explore effective deep learning methods to recognize succinylation sites. In this paper, we propose a multilane dense convolutional attention network, MDCAN-Lys. MDCAN-Lys extracts sequence information, physicochemical properties of amino acids, and structural properties of proteins using a three-way network, and it constructs feature space. For each sub-network, MDCAN-Lys uses the cascading model of dense convolutional block and convolutional block attention module to capture feature information at different levels and improve the abstraction ability of the network. The experimental results of 10-fold cross-validation and independent testing show that MDCAN-Lys can recognize more succinylation sites, which is consistent with the conclusion of the case study. Thus, it is worthwhile to explore deep learning-based methods for the recognition of succinylation sites.

show abstract

Section: Discussionmentioning

confidence: 99%

“…At present, more and more researchers have applied computational methods to the prediction of protein PTMs [9][10][11][12][13][14][15][16][17], RNA pseudouridine sites [18], and DNA methylation sites [19]. Machine learning-based methods have been widely used in the prediction of succinylation sites.…”

Section: Introductionmentioning

confidence: 99%

MDCAN-Lys: A Model for Predicting Succinylation Sites Based on Multilane Dense Convolutional Attention Network

et al. 2021

View full text Add to dashboard Cite

show abstract

“…Beyond just predicting Gene Ontology labels, studies have also focused on several other task-specific functional categories such as identifying specific enzyme functions 38 and potential post-translational modification sites 39 . These studies are a fundamental step towards developing novel proteins with specialized functions or modifying the efficacy of existing proteins as seen in the recent advances of DL in enzyme engineering 40 .…”

Section: Major Successes Of DLmentioning

confidence: 99%

Current progress and open challenges for applying deep learning across the biosciences

et al. 2022

View full text Add to dashboard Cite

Deep Learning (DL) has recently enabled unprecedented advances in one of the grand challenges in computational biology: the half-century-old problem of protein structure prediction. In this paper we discuss recent advances, limitations, and future perspectives of DL on five broad areas: protein structure prediction, protein function prediction, genome engineering, systems biology and data integration, and phylogenetic inference. We discuss each application area and cover the main bottlenecks of DL approaches, such as training data, problem scope, and the ability to leverage existing DL architectures in new contexts. To conclude, we provide a summary of the subject-specific and general challenges for DL across the biosciences.

show abstract

“…It is worth noting that with the development of deep learning and artificial intelligence technology, computers will likely become important tools to help us predict PTM sites. A novel computing framework DeepGSH (http://deepgsh.cancerbio.info/), which is based on depth learning and particle swarm optimization algorithms, was developed for the prediction of S-glutathionylation [50]. However, this system can only be applied to Homo sapiens and Mus musculus at present, and its accuracy needs to be verified.…”

Section: Plos Pathogensmentioning

confidence: 99%

S-glutathionylation proteome profiling reveals a crucial role of a thioredoxin-like protein in interspecies competition and cariogenecity of Streptococcus mutans

Zhang

Cheng

et al. 2020

PLoS Pathog

View full text Add to dashboard Cite

S-glutathionylation is an important post-translational modification (PTM) process that targets protein cysteine thiols by the addition of glutathione (GSH). This modification can prevent proteolysis caused by the excessive oxidation of protein cysteine residues under oxidative or nitrosative stress conditions. Recent studies have suggested that protein S-glutathionylation plays an essential role in the control of cell-signaling pathways by affecting the protein function in bacteria and even humans. In this study, we investigated the effects of Sglutathionylation on physiological regulation within Streptococcus mutans, the primary etiological agent of human dental caries. To determine the S-glutathionylated proteins in bacteria, the Cys reactive isobaric reagent iodoacetyl Tandem Mass Tag (iodoTMT) was used to label the S-glutathionylated Cys site, and an anti-TMT antibody-conjugated resin was used to enrich the modified peptides. Proteome profiling identified a total of 357 glutathionylated cysteine residues on 239 proteins. Functional enrichment analysis indicated that these Sglutathionylated proteins were involved in diverse important biological processes, such as pyruvate metabolism and glycolysis. Furthermore, we studied a thioredoxin-like protein (Tlp) to explore the effect of S-glutathionylation on interspecies competition between oral streptococcal biofilms. Through site mutagenesis, it was proved that glutathionylation on Cys41 residue of Tlp is crucial to protect S. mutans from oxidative stress and compete with S. sanguinis and S. gordonii. An addition rat caries model showed that the loss of S-glutathionylation attenuated the cariogenicity of S. mutans. Taken together, our study provides an insight into the S-glutathionylation of bacterial proteins and the regulation of oxidative stress resistance and interspecies competition.

show abstract

Deep learning based prediction of species-specific protein S-glutathionylation sites

Cited by 22 publications

References 20 publications

MDCAN-Lys: A Model for Predicting Succinylation Sites Based on Multilane Dense Convolutional Attention Network

MDCAN-Lys: A Model for Predicting Succinylation Sites Based on Multilane Dense Convolutional Attention Network

Current progress and open challenges for applying deep learning across the biosciences

S-glutathionylation proteome profiling reveals a crucial role of a thioredoxin-like protein in interspecies competition and cariogenecity of Streptococcus mutans

Contact Info

Product

Resources

About