Expression profiling of immune cells in systemic lupuserythematosus by single-cell RNA sequencing

Systemic lupus erythematosus(SLE)is a systemic autoimmune disease characterized by abnormal cellular and humoral immune responses and excessive autoantibody production.The precise pathologic mechanism of SLE remains elusive.The advent of single-cell RNA sequencing(scRNA-seq)enables unbiased analysis of the molecular differences of cell populations at the single-cell level.We used scRNA-seq to profile the transcriptomes of peripheral blood mononuclear cells from an SLE patient compared with a healthy control(HC).A total of 16,021 cells were analyzed and partitioned into 12 distinct clusters.The marker genes of each cluster and the four major immune cell types(B cells,CD4+T cells,CD8+T cells,myeloid cells,and NK cells)were determined.Moreover,several genes involved in antigen processing and presentation through MHCII were highly enriched.GO enrichment analyses revealed abnormal gene expression patterns and signaling pathways in SLE.Of note,pseudotime analysis revealed that there was a different lineage hierarchy in the peripheral blood mononuclear cells(PBMCs)of the SLE patient,indicating that the cell states were substantially altered under disease conditions.Our analysis provides a comprehensive map of the cell types and states of the PBMCs of SLE patients at the single-cell level for a better understanding of the pathogenesis,diagnosis,and treatment of SLE.

A Generation Method of Letter-Level Adversarial Samples

In recent years,with the rapid development of natural language processing,the security issues related to it have attracted more and more attention.Character perturbation is a common security problem.It can try to completely modify the input classification judgment of the target program without people’s attention by adding,deleting,or replacing several characters,which can reduce the effectiveness of the classifier.Although the current research has provided various methods of perturbation attacks on characters,the success rate of some methods is still not ideal.This paper mainly studies the sample generation of optimal perturbation characters and proposes a characterlevel text adversarial sample generation method.The goal is to use this method to achieve the best effect on character perturbation.After sentiment classification experiments,this model has a higher perturbation success rate on the IMDB dataset,which proves the effectiveness and rationality of this method for text perturbation and provides a reference for future research work.