ToxGIN: an In silico prediction model for peptide toxicity via graph isomorphism networks integrating peptide sequence and structure information DOI Creative Commons
Qi Yu, Zhixing Zhang, Guixia Liu

и другие.

Briefings in Bioinformatics, Год журнала: 2024, Номер 25(6)

Опубликована: Сен. 23, 2024

Abstract Peptide drugs have demonstrated enormous potential in treating a variety of diseases, yet toxicity prediction remains significant challenge drug development. Existing models for peptide largely rely on sequence information and often neglect the three-dimensional (3D) structures peptides. This study introduced novel model short prediction, named ToxGIN. The utilizes Graph Isomorphism Network (GIN), integrating underlying amino acid composition 3D ToxGIN comprises three primary modules: (i) Sequence processing module, converting sequences into nodes edges; (ii) Feature extraction utilizing GIN to learn discriminative features from (iii) Classification employing fully connected classifier prediction. performed well independent test set with F1 score = 0.83, AUROC 0.91, Matthews correlation coefficient 0.68, better than existing toxicity. These results validated effectiveness structural data using proposed can be freely accessible at https://github.com/cihebiyql/ToxGIN.

Язык: Английский

Exploring the impact of bioactive peptides from fermented Milk proteins: A review with emphasis on health implications and artificial intelligence integration DOI
Hosam M. Habib, Rania Ismail, Mahmoud Agami

и другие.

Food Chemistry, Год журнала: 2025, Номер unknown, С. 144047 - 144047

Опубликована: Март 1, 2025

Язык: Английский

Процитировано

2
Bioactive potential and storage behavior of low molecular mass peptides in Pilsner and IPA style craft beers DOI Creative Commons
Roberta Nogueira Pereira da Silva,

Angelica Priscila Parussolo Tonin,

Gabriela Soares Macello Ramos

и другие.

Frontiers in Food Science and Technology, Год журнала: 2025, Номер 5

Опубликована: Янв. 22, 2025

Beer, one of the most widely consumed alcoholic beverages globally, is typically produced from barley and hops, contains carbohydrates, proteins, vitamins, minerals, ethanol, bioactive phytochemicals such as phenolic compounds. However, knowledge protein content, particularly peptides in beer, remains limited. Given that beer production involves raw materials rich both proteins proteolytic enzymes, which may remain active throughout product’s shelf life, holds potential a source peptides. This study aimed to investigate presence di- tripeptides craft samples Pilsner IPA styles, after 3 or 6 months storage. LC-MS/MS analysis was performed using 46 Da neutral loss method collision-induced dissociation, followed by peptide bioactivity screening through BIOPEP database. Twelve tripeptides, with masses ranging 177 329 (m/z), were identified, exhibiting bioactivities dipeptidyl peptidase IV III inhibition, ACE antioxidative properties. These activities are associated reduced risk high blood pressure metabolic syndrome. After storage, intensity decreased but increased samples. beers, clear due added chill-proofing proteases, showed over time, whereas IPA, often hazy lacks exhibited levels. findings suggest beers benefit quicker consumption, while be better suited for longer storage maximize intake.

Язык: Английский

Процитировано

0
ToxDL 2.0: Protein toxicity prediction using a pretrained language model and graph neural networks DOI Creative Commons

Lin Zhu,

Yi Fang,

Shuting Liu

и другие.

Computational and Structural Biotechnology Journal, Год журнала: 2025, Номер unknown

Опубликована: Апрель 1, 2025

Assessing the potential toxicity of proteins is crucial for both therapeutic and agricultural applications. Traditional experimental methods protein evaluation are time-consuming, expensive, labor-intensive, highlighting requirement efficient computational approaches. Recent advancements in language models deep learning have significantly improved prediction, yet current often lack ability to integrate evolutionary structural information, which accurate assessment proteins. In this study, we present ToxDL 2.0, a novel multimodal model prediction that integrates information derived from pretrained AlphaFold2. 2.0 consists three key modules: (1) Graph Convolutional Network (GCN) module generating graph embeddings based on AlphaFold2-predicted structures, (2) domain embedding capturing representations, (3) dense combines these predict toxicity. After constructing comprehensive benchmark dataset, obtained results an original non-redundant test set (comprising pre-2022 sequences) independent (a holdout post-2022 sequences), demonstrating outperforms existing state-of-the-art methods. Additionally, utilized Integrated Gradients discover known toxic motifs associated with A web server publicly available at www.csbio.sjtu.edu.cn/bioinf/ToxDL2/.

Язык: Английский

Процитировано

0
Integrated computational approaches for advancing antimicrobial peptide development DOI

Yanpeng Fang,

Yeshuo Ma, Kunqian Yu

и другие.

Trends in Pharmacological Sciences, Год журнала: 2024, Номер 45(11), С. 1046 - 1060

Опубликована: Окт. 25, 2024

Язык: Английский

Процитировано

2
ToxGIN: an In silico prediction model for peptide toxicity via graph isomorphism networks integrating peptide sequence and structure information DOI Creative Commons
Qi Yu, Zhixing Zhang, Guixia Liu

и другие.

Briefings in Bioinformatics, Год журнала: 2024, Номер 25(6)

Опубликована: Сен. 23, 2024

Abstract Peptide drugs have demonstrated enormous potential in treating a variety of diseases, yet toxicity prediction remains significant challenge drug development. Existing models for peptide largely rely on sequence information and often neglect the three-dimensional (3D) structures peptides. This study introduced novel model short prediction, named ToxGIN. The utilizes Graph Isomorphism Network (GIN), integrating underlying amino acid composition 3D ToxGIN comprises three primary modules: (i) Sequence processing module, converting sequences into nodes edges; (ii) Feature extraction utilizing GIN to learn discriminative features from (iii) Classification employing fully connected classifier prediction. performed well independent test set with F1 score = 0.83, AUROC 0.91, Matthews correlation coefficient 0.68, better than existing toxicity. These results validated effectiveness structural data using proposed can be freely accessible at https://github.com/cihebiyql/ToxGIN.

Язык: Английский

Процитировано

1