Liver cancer patient survival prediction model construction method based on cell death related genes

The invention specifically discloses a hepatocellular carcinoma patient survival prediction model construction method based on cell death related genes. The method comprises the following steps: S1, constructing a preliminary hepatocellular carcinoma patient survival risk score prediction model; s2, a public database TCGA is used as a training set, and genes are expressed on the basis of differential expression related to three novel programmed cell death including cell autophagy, cell ferroptosis and pyroptosis; s3, determining a gene related to the lifetime through single-factor Cox regression analysis; s4, screening the genes related to the lifetime through multi-factor Cox regression analysis, and training to obtain a final hepatocellular carcinoma patient survival risk score prediction model; and S5, calculating a risk index according to the gene related expression quantity and the risk related coefficient, and analyzing and performing external verification. According to the method, the survival of the h