Fig. 2

The risk score model could effectively predict the survival of OSCC patients. a The forest plot of univariate Cox regression analysis of 26 hypoxia-related genes. HR: Hazard ratio, CI: confidence interval. *P < 0.05. b Determination of the tuning parameter lambda by using LASSO Cox regression analysis. The horizontal axis represents log (lambda), and the vertical axis represents partial likelihood deviation. The Lambda value which corresponds to the smallest partial likelihood deviance value is the optimal. That is, the optimal Lambda value after Log is taken below the dotted line, and the corresponding value above is the number of optimal genes. c Coefficient spectrum of LASSO Cox regression analysis. d The distribution of Risk Scores of the OSCC samples in the TCGA dataset. The red color represents high expression, and the green color represents low expression. e Violin diagram of the expression levels of four hypoxia-related genes in the high and low risk groups of TCGA samples. f Kaplan Meier survival curve of TCGA samples. The horizontal axis denotes time, the vertical axis denotes survival rate, and different groups are represented by different colors. g The distribution of Risk Scores of samples in the GSE65858. h Violin diagram of the expression levels of four hypoxia-related genes in the GSE65858 in the high and low risk groups. i Kaplan Meier survival curve of GSE65858