Prognosis analysis of methyl transferase-related factors.

Good weekend, everyone. Today I'd like to share with you an article published in frontier in oncology (if: 4.85) at the end of April 2020, which has a routine (typing screening marker prognostic model construction, simple 😊） But the focus is on methyltransferase related factors.methylation is a reversible modification, and the proteins and genes regulating this modification can be used as candidate targets for tumor therapy.however, the characteristics of methyltransferase related genes in gliomas are not clear.based on TCGA data and RNA sequencing data set of Chinese glioma genome map, this paper systematically analyzed the relationship between the expression of methyltransferase related genes and the prognosis of glioma patients by using common bioinformatics methods such as consistent clustering and Cox risk regression model, and constructed a prognostic marker related to methyl transfer gene.on the whole, it is a bioinformatics article full of routines.Clinical and biological signals of a methyltransferase related signature in diffuse glioma: clinical and biological significance of methyltransferase related features in diffuse glioma.methyltransferase related genes were obtained from msigdb database.2. Glioma was classified based on methyl metastasis related genes, and patients were divided into two distinct categories based on the consistency clustering of 50 methyltransferase related genes with the highest change degree in TCGA data (Fig. 1a) C) There are significant differences in prognosis, grade, IDH mutation and other clinical and molecular characteristics between the two groups. The thermogram of the two clusters is shown in Fig. 1D, and the prognosis is shown in Fig. 1E. These results indicate that the genes related to methyl metastasis are related to the malignant degree and prognosis of diffuse glioma.Figure 1. Classification of gliomas based on methyltransferase related genes3. Construction of prognostic markers of methyltransferase related genes in diffuse gliomas. Based on univariate Cox regression analysis, 65 prognostic related genes in TCGA data were identified, and the dimensionality of genes was reduced by lasso regression algorithm (Fig. 2a, b). Finally, the prognostic characteristics of 12 genes were identified.the risk score was calculated according to the expression level and regression coefficient of patients (Fig. 2B, c), and the patients were divided into high-risk group and low-risk group with the median as the threshold.there were differences in age, grade, IDH mutation status and other clinical and molecular characteristics between the two groups.similar differences were observed in the CGGA validation data between the two groups of patients.
Figure 2. Construction of gene prognostic markers 4. Prognostic risk score is related to the pathological characteristics of glioma. Through the study on the relationship between risk score and pathological characteristics of patients, it can be observed that there are significant differences in risk scores of patients when patients are grouped according to different clinical and molecular characteristics (IDH status, who score, subtype, etc., FIG. 3a-j).subsequently, the specificity and sensitivity of the risk score in predicting pathological features were evaluated by ROC curve (category, IDH status, 1p / 9q status, figure 3K – m).the risk score can predict the sample category, IDH status and 1p / 9q status in TCGA and CGGA datasets.Figure 3. The prognostic risk score is related to the prognosis of glioma. 5. The prognosis risk score is related to the prognosis of glioma. According to the commonly used clinical and molecular characteristics such as grade, IDH mutation status, Mgmt promoter status, it is found that in patients with specific molecular characteristics, the risk score can also divide patients into high-risk group and high-risk group In the difference. The prognostic risk score is an independent prognostic marker. Based on univariate and multivariate Cox analysis, whether the risk score is an independent prognostic indicator is evaluated. as shown in Fig. 5a, diagnostic age, WHO classification, risk score and other factors were statistically correlated with overall survival (OS) of glioma patients. among them, diagnostic age and risk score are independent prognostic indicators of OS in patients with glioma in TCGA data set (Table 3, figure 5a). compared with age and grade of traditional factors, risk score was more accurate in predicting 3-year and 5-year survival in TCGA data set (Fig. 5b, c). similar results can be observed in CGGA data. Figure 5. Prognostic risk score is an independent prognostic marker. 7. Biological characteristics and pathway analysis of prognostic markers. Based on Pearce correlation analysis, genes expressing positive or negative correlation with risk score were identified. Go and KEGG enrichment analysis were performed for these two types of genes. The results are shown in Figure 6. These pathways or biological processes are considered to be related to the prognosis of methyl metastasis or glioma Off. Figure 6. Biological characteristics and pathway analysis of prognostic markers. the researchers used 50 highly variable expression of methyltransferase related genes to cluster the patients. There were significant differences in clinical and molecular characteristics between the two groups of patients, indicating that methyl transfer related genes are related to the malignant degree and prognosis of diffuse glioma. then, through the univariate Cox regression analysis and lasso model, we reduced the dimension of genes related to methyl metastasis, and constructed the prognostic marker of 12 genes, which is an independent prognostic marker, and is closely related to the pathological and prognostic characteristics of glioma. although this article has some simple routines, it is very fluent. There is a relationship between each step and the previous step, and it is very complete. It is worth learning.