Tripterygium glycosides tablet(TGT),the classical commercial drug of Tripterygium wilfordii Hook.F.has been effectively used in the treatment of rheumatoid arthritis,nephrotic syndrome,leprosy,Behcet's syndrome,leprosy reaction and autoimmune hepatitis.However,due to its narrow and limited treatment window,TGT-induced organ toxicity(among which liver injury accounts for about 40％of clinical reports)has gained increasing attention.The present study aimed to clarify the cellular and molecular events underlying TGT-induced acute liver injury using single-cell RNA sequencing(scRNA-seq)technology.The TGT-induced acute liver injury mouse model was constructed through short-term TGT exposure and further verified by hematoxylin-eosin staining and liver function-related serum indicators,including alanine aminotransferase,aspartate aminotransferase,alkaline phosphatase and total bilirubin.Using the mouse model,we identified 15 specific subtypes of cells in the liver tissue,including endothelial cells,hepatocytes,cholangiocytes,and hepatic stellate cells.Further analysis indicated that TGT caused a significant inflammatory response in liver endothelial cells at different spatial locations;led to marked inflammatory response,apoptosis and fatty acid metabolism dysfunction in hepatocytes;activated he-patic stellate cells;brought about the activation,inflammation,and phagocytosis of liver capsular macrophages cells;resulted in immune dysfunction of liver lymphocytes;disturbed the intercellular crosstalk in liver microenvironment by regulating various signaling pathways.Thus,these findings elaborate the mechanism underlying TGT-induced acute liver injury,provide new insights into the safe and rational applications in the clinic,and complement the identification of new biomarkers and ther-apeutic targets for liver protection.

Individual differences of P300 potentials lead to that a large amount of training data must be collected to construct pattern recognition models in P300-based brain-computer interface system, which may cause subjects' fatigue and degrade the system performance. TrAdaBoost is a method that transfers the knowledge from source area to target area, which improves learning effect in the target area. Our research purposed a TrAdaBoost-based linear discriminant analysis and a TrAdaBoost-based support vector machine to recognize the P300 potentials across multiple subjects. This method first trains two kinds of classifiers separately by using the data deriving from a small amount of data from same subject and a large amount of data from different subjects. Then it combines all the classifiers with different weights. Compared with traditional training methods that use only a small amount of data from same subject or mixed different subjects' data to directly train, our algorithm improved the accuracies by 19.56% and 22.25% respectively, and improved the information transfer rate of 14.69 bits/min and 15.76 bits/min respectively. The results indicate that the TrAdaBoost-based method has the potential to enhance the generalization ability of brain-computer interface on the individual differences.
Algorithms ; Brain-Computer Interfaces ; Discriminant Analysis ; Electroencephalography ; Event-Related Potentials, P300 ; Humans ; Support Vector Machine