Expression and immunological infiltration of TDP-43-related genes in idiopathic pulmonary fibrosis based on bioinformatics analysis
[Objective]The.relationship between expression of TDP-43 related gene and immune infiltration in idiopathic pul-monary fibrosis(IPF)was analyzed based on bioinformatics.[Method]Obtain the transcriptome data in database from Gene Expression Omnibus(GEO),through the correlation analysis with TDP-43(|correlation coefficient|＞0.75)for TDP-43 related genes,and the analysis of the differences between the TDP-genetic differences associated with 43.Four machine learn-ing algorithms were used to establish models to screen candidate targets for predicting the risk of IPF.Single sample gene set enrichment analysis(ssGSEA)was used to analyze the immune cell infiltration of the candidate targets.Based on the expres-sion of candidate targets,the training set was divided into 3 subgroups.Gene set variation analysis was used to evaluate the en-richment of metabolic pathways among the three groups.In addition,the differences in immune cell infiltration among the three subgroups were compared.[Result]A total of 58 differential genes related to TDP-43 were identified.Among the four ma-chine learning algorithms,the SVM optimal algorithm was used to construct the model.TIMM17A,RCOR1,HTATSF1,SENP5 and GNS were identified as potential diagnostic biomarkers,and the area under the ROC curve of internal and external valida-tion was 0.955 and 0.888,respectively.In addition,a nomogram was developed to predict the risk.There was a difference in the degree of immune cell infiltration between high and low expression of the five potential markers.Metabolic pathways and im-mune infiltration were different in the three subgroups.[Conclusion]TIMM17A,RCOR1,HTATSF1,SENP5,and GNS may be used as markers for the diagnosis of IPF,and TDP-43 related genes may affect the immune infiltration of IPF.
bioinformaticsidiopathic pulmonary fibrosisGEOimmune infiltrationdiagnosisbiomarkersTDP-43 related genesmachine learning
NETL
NSTL
万方数据
