Fig. 4

Comparison of regression parameter ranges. The regression parameter ranges of linear (a), quadratic (b), and cubic polynomial models (c), as well as the quadratic polynomial models with 10 correlated gene pairs (d). For the linear model, the first parameter represents the constant term of each regression function, and other parameters are arranged according to the importance of the corresponding variables. Each variable stands for the expression level of a gene, and its importance is evaluated by the absolute value of the correlation coefficient with cell death. For the quadratic model, the parameters are arranged in the order of constant term, the parameters of the linear variables, and the parameters of quadratic variables. The ranks of the parameters of both linear and quadratic variables can refer to that of the linear model. The parameters of the cubic model are shown similarly