| ObjectiveIn this study,two datasets,GSE37265 and GSE8056,were screened by bioinformatics methods and eventually screened for early diagnostic markers of trauma recovery at the level of trauma healing after differential gene screening.Method1.Open the GEO database to filter transcriptome sequencing data for skin lesions with the keywords skin and wound,and filter for gene sequencing datasets containing normal skin tissue,and traumatised skin tissue.2.Organize and visualize the data using the GEOquery package,limma package,umap package,cluster package,Heatmap package and ggplot2 package in R language(version 4.1.1).3.The two datasets,GSE37265 and GSE8056,were processed as described above,and the screened differential genes were analysed to identify genes that differed in both datasets.4.The common differential genes were subjected to GO and KEGG analysis to identify the associated molecular functions,biological processes and cellular components,and associated pathways in skin trauma.5.The co-expressed differential genes were subjected to protein interaction network construction using the STRING database and the network relationships between genes were derived.6.The network relationships between genes from the previous step are calculated using the MOCDE package using cytoscape software,resulting in trauma-associated key genes and identifying the final target genes for study.7.The trauma tissue from outpatient trauma dressing changes was collected as samples and divided into good healing group and poor healing group according to the trauma healing condition,RNA was extracted by trizol method,and RT-PCR was used to detect the expression of differential genesResults1.The UAMP plots and PCA principal component analysis plots of the two datasets GSE37265,GSE8056 show that the samples are well differentiated in the two data sets,and the volcano plot and heat map show that the samples have significant and abundant differentially expressed genes.2.577 genes were screened in dataset GSE8056 that met the threshold of log2(FC)|>1&p.adj<0.05,of which 313 were highly expressed and 264 were lowly expressed.The number of genes with log2(FC)|>1&p.adj<0.05 threshold in dataset GSE37265 was 682,the number of high expression was 579 and the number of low expression was 103.3.The 577 differential genes screened against the 682 differential genes visualised by the Venn diagram showed that a total of 140 genes were significantly differentially expressed in both sets of data.4.Ten key node genes of TIMP1,IL6,EPSTI1,CCL2,CXCL1,IL1-β,MMP3,MMP9,MMP1 can be derived from the STRING online data by calculating the protein interaction network of the above 140 differential genes and further calculation by Cytohubba using cytoscape software.5.The PCR results showed that relative to the good prognosis,the expression level of TIMP1 was 1.5-fold higher,IL6 was 14.43-fold higher,EPSTI1 was 1.09-fold higher,CCL2 was 1.56-fold higher,CXCL1 was 717.78-fold higher,MMP3 was 3.04-fold higher,MMP9 was 5.30-fold higher,MMP1 was 9.44-fold higher,CXCL8 was 3.04-fold higher,MMP9 was 5.30-fold higher,MMP1 was 9.44-fold higher,CXCL2 was 1.56-fold higher,CXCL1 was 717.78-fold higher.The differences were statistically significant,with P values<0.05.ConclusionsBy analyzing the skin trauma-related sequencing results in the datasets GSE37265,GSE8056,we predicted an increased expression of 10 genes in poorly healed trauma.The results of PCR assay by collecting clinical samples from wounds with different healing conditions support the results of previous data analysis.TIMP1,IL6,EPSTI1,CCL2,CXCL1,IL1-β,MMP3,MMP9,MMP1 can be used as potential biological indicators to monitor wound healing... |
PDF Full Text Request