Versatile Framework for Drug-Target Interaction Prediction by Considering Domain-Specific Features.

Predicting drug-target interactions (DTIs) is one of the crucial tasks in drug discovery, but traditional wet-lab experiments are costly and time-consuming. Recently, deep learning has emerged as a promising tool for accelerating DTI prediction due to its powerful performance. However, the models trained on limited known DTI data struggle to generalize effectively to novel drug-target pairs. In this work, we propose a strategy to train an ensemble of models by capturing both domain-generic and domain-specific features (E-DIS) to learn diverse domain features and adapt them to out-of-distribution data. Multiple experts were trained on different domains to capture and align domain-specific information from various distributions without accessing any data from unseen domains. E-DIS provides a comprehensive representation of proteins and ligands by capturing diverse features. Experimental results on four benchmark data sets in both in-domain and cross-domain settings demonstrated that E-DIS significantly improved model performance and domain generalization compared to existing methods. Our approach presents a significant advancement in DTI prediction by combining domain-generic and domain-specific features, enhancing the generalization ability of the DTI prediction model.

Antibacterial activity and mechanism of luteolin isolated from Lophatherum gracile Brongn. against multidrug-resistant Escherichia coli.

Infections caused by multidrug-resistant (MDR) bacteria have become a major challenge for global healthcare systems. The search for antibacterial compounds from plants has received increasing attention in the fight against MDR bacteria. As a medicinal and edible plant, Lophatherum gracile Brongn. (L. gracile) has favorable antibacterial effect. However, the main antibacterial active compound and its antimicrobial mechanism are not clear. Here, our study first identified the key active compound from L. gracile as luteolin. Meanwhile, the antibacterial effect of luteolin was detected by using the broth microdilution method and time-kill curve analysis. Luteolin can also cause morphological structure degeneration and content leakage, cell wall/membrane damage, ATP synthesis reduction, and downregulation of mRNA expression levels of sulfonamide and quinolones resistance genes in multidrug-resistant Escherichia coli (MDR E. coli). Furthermore, untargeted UPLC/Q-TOF-MS-based metabolomics analysis of the bacterial metabolites revealed that luteolin significantly changed riboflavin energy metabolism, bacterial chemotaxis cell process and glycerophospholipid metabolism of MDR E. coli. This study suggests that luteolin could be a potential new food additive or preservative for controlling MDR E. coli infection and spread.

Analysis of authorized coccidiostats in chicken feces and environmental water by ultra-performance liquid chromatography-tandem mass spectrometry based on dispersive solid-phase extraction and lyophilization.

A simple, sensitive, and efficient method based on ultra-performance liquid chromatography-tandem mass spectrometry (UPLC-MS/MS) was developed for the determination of 8 coccidiostats in chicken feces and environmental water (including sewage, pond water, and lake water) surrounding the farm. Target analytes in chicken feces were extracted with 2% acetic acid in acetonitrile solution, followed by a dispersive solid-phase extraction (DSPE) cleanup step using the mixture of PSA and C18 adsorbents. Environmental water samples were pretreated using a lyophilization approach. Analysis was carried out on a UPLC-MS/MS with the combination of methanol and 0.1% formic acid aqueous solution as the mobile phase under multiple reaction monitoring in positive and negative ionization modes. Results showed that 8 coccidiostats were linear with correlation coefficients higher than 0.99. Method validation was performed using fortified samples, reaching satisfactory recoveries of 75.9%-97.8% in chicken feces and 71.9%-108.2% in environmental water. Limits of detection for 8 analytes in chicken feces and environmental water were 0.03∼2 µg/kg and 0.005∼1 µg/L, respectively. Matrix effects were calculated and strong signal suppression (>50%) for some coccidiostats was observed. The developed method was successfully applied to analyze coccidiostats in chicken feces and environmental water collected from local chicken farms.