Discovery of novel VEGFR2 inhibitors against non-small cell lung cancer based on fingerprint-enhanced graph attention convolutional network DOI Creative Commons

Zixiao Wang,

Lili Sun,

Yu Xu

и другие.

Journal of Translational Medicine, Год журнала: 2024, Номер 22(1)

Опубликована: Дек. 3, 2024

Despite the proven inhibitory effects of drugs targeting vascular endothelial growth factor receptor 2 (VEGFR2) on solid tumors, including non-small cell lung cancer (NSCLC), development anti-NSCLC solely VEGFR2 still faces risks such as off-target and limited efficacy. This study aims to develop a novel fingerprint-enhanced graph attention convolutional network (FnGATGCN) model for predicting activity drugs. Employing multimodal fusion strategy, integrates feature extraction layer that comprises molecular fingerprint extraction. The performance evaluation results indicate exhibits high accuracy stability in activity. Moreover, we explored relationship between features biological through visualization analysis, thus improving interpretability approach. Utilizing this model, screened ZINC database conducted high-precision docking, leading identification 11 potential active molecules. Subsequently, dynamics simulations free energy calculations were performed. demonstrate all aforementioned molecules can stably bind under dynamic conditions. Among short-listed compounds, top six exhibited satisfactory against A549 cells. Especially, compound Z-3 displayed with IC50 values 0.88 μM, anti-proliferative cells 4.23 ± 0.45 μM. approach combines advantages target-based phenotype-based screening, facilitating rapid efficient candidate compounds dual lines. It provides new insights methods Furthermore, further tests revealed Z1-Z3 Z6 manifested relatively strong antiproliferative activities NCI-H23 NCI-H460, low toxicity towards GES-1. hit promising candidates inhibitors NSCLC.

Язык: Английский

Isolation, Virtual Screening, and Evaluation of Hazelnut-Derived Immunoactive Peptides for the Inhibition of SARS-CoV-2 Main Protease DOI
Xiaoting Liu, Shuo Sun, Jiale Liu

и другие.

Journal of Agricultural and Food Chemistry, Год журнала: 2024, Номер 72(20), С. 11561 - 11576

Опубликована: Май 13, 2024

The aim of this study is to validate the activity hazelnut (Corylus avellana L.)-derived immunoactive peptides inhibiting main protease (Mpro) SARS-CoV-2 and further unveil their interaction mechanism using in vitro assays, molecular dynamics (MD) simulations, binding free energy calculations. In general, enzymatic hydrolysis components, especially weight < 3 kDa, possess good immune as measured by proliferation ability mouse splenic lymphocytes phagocytic peritoneal macrophages. Over 866 unique peptide sequences were isolated, purified, then identified nanohigh-performance liquid chromatography/tandem mass spectrometry (NANO-HPLC-MS/MS) from protein hydrolysates, but Trp-Trp-Asn-Leu-Asn (WWNLN) Trp-Ala-Val-Leu-Lys (WAVLK) particular are found increase cell viability capacity RAW264.7 macrophages well promote secretion cytokines nitric oxide (NO), tumor necrosis factor-α (TNF-α), interleukin-1β (IL-1β). Fluorescence resonance transfer assay elucidated that WWNLN WAVLK exhibit excellent inhibitory potency against Mpro, with IC50 values 6.695 16.750 μM, respectively. Classical all-atom MD simulations show hydrogen bonds play a pivotal role stabilizing complex conformation protein–peptide interaction. Molecular Mechanics/Generalized Born Surface Area (MM/GBSA) calculation indicates has lower Mpro than WAVLK. Furthermore, adsorption, distribution, metabolism, excretion, toxicity (ADMET) predictions illustrate favorable drug-likeness pharmacokinetic properties compared This provides new understanding immunomodulatory hydrolysates sheds light on inhibitors targeting Mpro.

Язык: Английский

Процитировано

4
Large-scale Deep Learning Identifies the Antiviral Potential of PKI-179 and MTI-31 Against Coronaviruses DOI Creative Commons
Demi van der Horst, Madalina E. Carter-Timofte,

Adeline Danneels

и другие.

Antiviral Research, Год журнала: 2024, Номер 231, С. 106012 - 106012

Опубликована: Сен. 25, 2024

Язык: Английский

Процитировано

3
Hybrid intelligence for environmental pollution: biodegradability assessment of organic compounds through multimodal integration of graph attention networks and QSAR models DOI

Abbas Salimi,

Jin Yong Lee

Environmental Science Processes & Impacts, Год журнала: 2025, Номер unknown

Опубликована: Янв. 1, 2025

Computational methods are crucial for assessing chemical biodegradability, given their significant impact on both environmental and human health. Organic compounds that not biodegradable can persist in the environment, contributing to pollution. Our novel approach leverages graph attention networks (GATs) incorporates node edge attributes biodegradability prediction. Quantitative Structure-Activity Relationship (QSAR) models using two-dimensional descriptors alongside weighted average stacking approaches were employed generate ensemble models. The GAT demonstrated a stable function generally higher specificity validation set compared convolutional network, although definitive superiority is challenging establish owing overlapping standard deviations. However, sensitivities tended decrease with potential performance overlap interval intersection. Ensemble learning enhanced several metrics individual base models, combination of extreme Gradient Boosting achieving highest precision specificity. Combining random forest may be preferable accurately predicting molecules, whereas suitable prioritizing correct classification nonbiodegradable substances. Important descriptors, such as SpMax1_Bh(m) SAscore, identified at least two QSAR Despite inherent complexities, ease implementation depends factors data availability, domain knowledge. Assessing organic essential reducing impact, risks, ensuring regulatory compliance, promoting sustainable development, supporting effective pollution remediation. It assists making informed decisions about use, waste management, protection.

Язык: Английский

Процитировано

0
Discovery of novel VEGFR2 inhibitors against non-small cell lung cancer based on fingerprint-enhanced graph attention convolutional network DOI Creative Commons

Zixiao Wang,

Lili Sun,

Yu Xu

и другие.

Journal of Translational Medicine, Год журнала: 2024, Номер 22(1)

Опубликована: Дек. 3, 2024

Despite the proven inhibitory effects of drugs targeting vascular endothelial growth factor receptor 2 (VEGFR2) on solid tumors, including non-small cell lung cancer (NSCLC), development anti-NSCLC solely VEGFR2 still faces risks such as off-target and limited efficacy. This study aims to develop a novel fingerprint-enhanced graph attention convolutional network (FnGATGCN) model for predicting activity drugs. Employing multimodal fusion strategy, integrates feature extraction layer that comprises molecular fingerprint extraction. The performance evaluation results indicate exhibits high accuracy stability in activity. Moreover, we explored relationship between features biological through visualization analysis, thus improving interpretability approach. Utilizing this model, screened ZINC database conducted high-precision docking, leading identification 11 potential active molecules. Subsequently, dynamics simulations free energy calculations were performed. demonstrate all aforementioned molecules can stably bind under dynamic conditions. Among short-listed compounds, top six exhibited satisfactory against A549 cells. Especially, compound Z-3 displayed with IC50 values 0.88 μM, anti-proliferative cells 4.23 ± 0.45 μM. approach combines advantages target-based phenotype-based screening, facilitating rapid efficient candidate compounds dual lines. It provides new insights methods Furthermore, further tests revealed Z1-Z3 Z6 manifested relatively strong antiproliferative activities NCI-H23 NCI-H460, low toxicity towards GES-1. hit promising candidates inhibitors NSCLC.

Язык: Английский

Процитировано

0