Screening Of Diagnostic Marker And Therapeutic Drugs For The Progression Of Non-alcoholic Fatty Liver Disease Based On Bioinformatics

Objective NAFLD affects about a quarter of the global population.The diagnosis and treatment of the progression of non-alcoholic fatty liver disease has become an urgent medical problem.There is an urgent need to find biomarkers and drugs for the treatment of NAFLD.Methods The transcriptome sequencing data of NAFLD tissues at different stages were obtained from GEO database,and the data was GSE130970.Firstly,Limma package was used to identify differentially expressed genes in different stages of the disease to determine the biological functions and pathways on NAFLD.Next,Lasso regression model and random forest model were used to further analyze the key genes,and immune infiltration scoring algorithm was used to screen candidate genes.ROC curve,nomogram,calibration map and decision map were used to evaluate the predictive ability of candidate genes.Finally,the feasibility of candidate genes in independent disease datasets(GSE33814 and GSE89632)was evaluated according to ROC curve.dp GSEA analysis was used to screen candidate drugs for the treatment of NAFLD,and KEGG enrichment analysis was used to rank the candidate drugs.Finally,a cell model of NAFLD was established,and the levels of TG,ALT,and AST were detected to observe the changes before and after treatment and to determine the potential therapeutic effects of candidate drugs.Results we identified the differential genes of the control group-NASH and the control group-Fibrosis group,and obtained 254 differential genes and 274 differential genes,of which 117 genes appeared in the two groups at the same time.The ten characteristic genes were selected by Lasso method,and the nine genes were selected by random forest analysis.A total of 15 disease-related genes were obtained by the two methods.Combined with differential analysis and correlation analysis,the naive CD4~+T cells were related to the progression of NAFLD.Combined with Spearman correlation analysis and gene expression levels,IL32 gene and MYO16 gene could be used as candidate diagnostic markers for the progression of NAFLD.The prediction model of IL32 gene and MYO16 gene was established by ROC curve,nomogram,calibration curve and decision chart.The results showed that IL32 gene and MYO16 gene could effectively predict the different stages of NAFLD progression,which was validated in independent datasets(GSE33814 and GSE89632).The feasibility of IL32 gene and MYO16 gene as markers for the diagnosis of NAFLD progression was confirmed.After drug perturbation enrichment analysis,There were seven drugs or compounds were found to have potential therapeutic effects on the progression of NAFLD.KEGG analysis showed that Indoximod was related to NAFLD.Cell experiments were performed to verify that the addition of Indoximod to NAFLD cell model could significantly reverse the increase of triglyceride,alanine transaminase and aspartate transaminase induced by palmitic acid in L-02 cells.Conclusion IL32 gene and MYO16 gene can be used as diagnostic markers for the progression of NAFLD,and Indoximod has a potential therapeutic effect on the progression of NAFLD.The results of this study will provide theoretical support for the development and treatment of NAFLD.