Binding Affinity Prediction and Pesticide Screening against Phytophthora sojae Using a Heterogeneous Interaction Graph Attention Network–Based Model DOI

Youxu Dai,

Aiping Han, Hui-Jun Ma

и другие.

Journal of Chemical Information and Modeling, Год журнала: 2025, Номер unknown

Опубликована: Фев. 26, 2025

Phytophthora root and stem rot in soybeans results substantial economic losses worldwide. In this study, a machine learning model based on heterogeneous interaction graph attention network was constructed. The PDBbind data set, comprising 13,285 complexes with experimental pKa or pKi values, utilized to train evaluate the model, which subsequently employed screen candidate compounds against chitin synthase of sojae (PsChs1) Traditional Chinese Medicine Systems Pharmacology database, 14,249 compounds. High-scoring were docked PsChs1 protein using Discovery Studio, their energies evaluated. Molecular dynamic simulations spanning 50 ns performed GROMACS explore stability complexes, trajectory analysis conducted root-mean-square deviations, hydrogen bonds, radius gyration, MMPBSA binding free energy, modes analyzed. MOL011832 MOL011833 identified as potential pesticides, both present herb Schizonepeta through database retrieval. inhibitory effects an ethanol extract P. explored confirmed biological experiments. Overall, study proves feasibility high efficiency pesticide discovery neural network–based models.

Язык: Английский

Nitrogen Fertilization Coupled with Zinc Foliar Applications Modulate the Production, Quality, and Stress Response of Sideritis cypria Plants Grown Hydroponically Under Excess Copper Concentrations DOI Creative Commons
Nikolaos Tzortzakis, Giannis Neofytou, Antonios Chrysargyris

и другие.

Plants, Год журнала: 2025, Номер 14(5), С. 691 - 691

Опубликована: Фев. 24, 2025

The demand for medicinal and aromatic plants (MAPs) has grown significantly in recent years, due to their therapeutic value. Among these, Sideritis cypria Post is a promising yet under-evaluated species. Existing research assessing the effects of nitrogen (N) fertilization, zinc (Zn) foliar applications, toxic copper (Cu) concentrations often overlooks MAPs such as S. cypria. Additionally, interactions among these parameters, well combined roles plant physiology secondary metabolite biosynthesis, have be fully elucidated. In this study, hydroponically were cultivated using nutrient solutions (NSs) with different N (75, 150, 300 mg L−1) Cu (5 100 μM) levels, spraying (0 1.74 mM Zn), evaluate growth, mineral uptake, metabolites production stress response. levels at 75 150 L−1 resulted increased dry matter content, whereas fresh biomass was preserved. Foliar Zn applications enhanced chlorophylls antioxidants, contingent upon NS. Increased accumulation observed via increase NS, while its uptake moderate levels. Excess stimulated accumulation, reduction low high lipid peroxidation (MDA) decreased both MDA hydrogen peroxide, Low-to-moderate NS can applied under excess without compromising yield, quality, safety plants, modulate response metabolites. These results may utilized optimizing management strategies cultivation MAPs, contributing conservation efforts by supporting endemic species like cypria, considering potential benefits Cu-contaminated conditions.

Язык: Английский

Процитировано

0

Binding Affinity Prediction and Pesticide Screening against Phytophthora sojae Using a Heterogeneous Interaction Graph Attention Network–Based Model DOI

Youxu Dai,

Aiping Han, Hui-Jun Ma

и другие.

Journal of Chemical Information and Modeling, Год журнала: 2025, Номер unknown

Опубликована: Фев. 26, 2025

Phytophthora root and stem rot in soybeans results substantial economic losses worldwide. In this study, a machine learning model based on heterogeneous interaction graph attention network was constructed. The PDBbind data set, comprising 13,285 complexes with experimental pKa or pKi values, utilized to train evaluate the model, which subsequently employed screen candidate compounds against chitin synthase of sojae (PsChs1) Traditional Chinese Medicine Systems Pharmacology database, 14,249 compounds. High-scoring were docked PsChs1 protein using Discovery Studio, their energies evaluated. Molecular dynamic simulations spanning 50 ns performed GROMACS explore stability complexes, trajectory analysis conducted root-mean-square deviations, hydrogen bonds, radius gyration, MMPBSA binding free energy, modes analyzed. MOL011832 MOL011833 identified as potential pesticides, both present herb Schizonepeta through database retrieval. inhibitory effects an ethanol extract P. explored confirmed biological experiments. Overall, study proves feasibility high efficiency pesticide discovery neural network–based models.

Язык: Английский

Процитировано

0