A deep learning based multi-model approach for predicting drug-like chemical compound's toxicity.

Ensuring the safety and efficacy of chemical compounds is crucial in small-molecule drug development. In the later stages of drug development, toxic compounds pose a significant challenge, losing valuable resources and time. Early and accurate prediction of compound toxicity using deep learning models offers a promising solution to mitigate these risks during drug discovery. In this study, we present the development of several deep-learning models aimed at evaluating different types of compound toxicity, including acute toxicity, carcinogenicity, hERG_cardiotoxicity (the human ether-a-go-go related gene caused cardiotoxicity), hepatotoxicity, and mutagenicity. To address the inherent variations in data size, label type, and distribution across different types of toxicity, we employed diverse training strategies. Our first approach involved utilizing a graph convolutional network (GCN) regression model to predict acute toxicity, which achieved notable performance with Pearson R 0.76, 0.74, and 0.65 for intraperitoneal, intravenous, and oral administration routes, respectively. Furthermore, we trained multiple GCN binary classification models, each tailored to a specific type of toxicity. These models exhibited high area under the curve (AUC) scores, with an impressive AUC of 0.69, 0.77, 0.88, and 0.79 for predicting carcinogenicity, hERG_cardiotoxicity, mutagenicity, and hepatotoxicity, respectively. Additionally, we have used the approved drug dataset to determine the appropriate threshold value for the prediction score in model usage. We integrated these models into a virtual screening pipeline to assess their effectiveness in identifying potential low-toxicity drug candidates. Our findings indicate that this deep learning approach has the potential to significantly reduce the cost and risk associated with drug development by expediting the selection of compounds with low toxicity profiles. Therefore, the models developed in this study hold promise as critical tools for early drug candidate screening and selection.

Ursodeoxycholic acid protects interstitial Cajal-like cells in the gallbladder from undergoing apoptosis by inhibiting TNF-α expression.

Hypomotility is a common symptom of gallstone disease, which is accompanied by a loss of interstitial Cajal-like cells (ICLCs) in the gallbladder. Ursodeoxycholic acid (UDCA) is widely used in treating gallstone disease, and has shown anti-apoptotic and anti-inflammatory effects apart from its ability to dissolve gallstones. In this study, we investigated the anti-apoptotic and anti-inflammatory effects of UDCA on ICLCs in guinea pigs with gallstones. Guinea pigs were fed a high-cholesterol diet for 8 weeks to induce the formation of gallstones. A group of animals was administered UDCA (50 mg·kg-1·d-1, ig) simultaneously. At the end of 8 weeks, the animals were euthanized with anesthesia, cholecystectomy was performed immediately and gallbladder was collected for further analysis. We showed that in the model group the contractility of gallbladder muscle strips in response to both acetylcholine (ACh) and CCK-8 was severely impaired, which was significantly improved by UDCA administration. Furthermore, UDCA administration significantly reduced the apoptotic ratio of ICLCs, based on the observation of co-localization imaging of apoptotic cells and c-kit-positive cells. Western blotting analysis and real-time PCR results revealed that the TNF-α/Caspase8/Caspase3 pathway was suppressed in the UDCA-treated animals, confirming the anti-apoptotic effect of UDCA in the gallbladder. The H&E staining showed that UDCA administration significantly attenuated inflammatory cell infiltration in the gallbladder wall. In conclusion, UDCA can protect ICLCs in the gallbladder from undergoing apoptosis by inhibiting the TNF-α/Caspase8/caspase3 pathway.