| ObjectiveAge related macular degeneration mostly occurs in people over 50 years old,and its incidence rate increases with age,which is an important cause of blindness in the elderly.Cuprotosis is a newly discovered form of cell death,which seems to lead to the progress of various diseases.Therefore,this study aims to explore the diagnosis gene and treatment target of Cuprotosis in age-related macular degeneration,and build a model for verification.MethodFour gene expression data sets,GSE50195,GSE76923,GSE103060 and GSE125564,were selected from the GEO database to analyze the expression profile and immune characteristics of copper death regulator in age-related macular degeneration.45 age-related macular degeneration samples were used to explore molecular clusters based on Cuprotosis related genes and related immune cell infiltration.WGCNA algorithm was used to identify specific differentially expressed genes.Then,the best machine model is selected by comparing the performance of random forest model,support vector machine model,generalized linear model and XGB.The accuracy of the machine learning model is analyzed with an additional data set through nomographs,calibration curves,and decision curves.ResultFour age-related macular degeneration datasets from the GeneExpression Omnibus(GEO)database were combined.After eliminating the batch effect,the disordered copper death related genes were identified between age-related macular degeneration and the normal control group.According to the different expression of Cuprotosis gene,the group of age-related macular degeneration was divided into two different molecular clusters.The analysis of immune infiltration showed that there was significant immune difference between different clusters.Cluster2 is characterized by high immune score and relatively high level of immune infiltration.The support vector machine model shows the best learning performance and has a high area under the curve(AUC=1.000).Finally,a support vector machine model of the top five genes was constructed,which showed satisfactory results in an additional validation group data set(GSE76237).The analysis of nomogram,calibration curve and decision curve also proved the accuracy of age-related macular degeneration model learning.ConclusionThis study systematically explained the complex relationship between Cuprotosis and age-related macular degeneration,confirmed the involvement of immune mechanism in the process of Cuprotosis in AMD,and obtained new diagnostic genes and therapeutic targets for AMD. |
PDF Full Text Request