Table 2.
Akaike information criterion (AIC) of the various regression models
| Poisson | Zt- Poisson | Negative Binomial | Zt-Negative Binomial | |
|---|---|---|---|---|
| BLCA | 50439.28 | 49991.30 | 42766.17 | 40965.04 |
| BRCA | 93502.18 | 93343.70 | 60491.40 | 58701.48 |
| COAD | 55069.59 | 54675.87 | 40495.83 | 38459.15 |
| HNSC | 91759.62 | 91608.29 | 60737.90 | 59131.56 |
| KIRC | 49696.06 | 49152.82 | 38404.63 | 36484.36 |
| LUAD | 140194.62 | 140168.95 | 79081.09 | 78048.05 |
| LUSC | 79647.23 | 79409.43 | 55650.58 | 54129.60 |
| OV | 53288.73 | 52812.71 | 40753.37 | 38623.84 |
| SKCM | 126186.28 | 126133.21 | 70968.39 | 69622.20 |
| STAD | 63094.87 | 62754.31 | 48865.94 | 47114.14 |
| UCEC | 66024.20 | 65665.04 | 48584.07 | 46526.46 |
Note: The glm() function in R was used to fit the generalized linear model (GLM) of Poisson distribution. The glm.nb() function in the R package of MASS was used to fit the GLM of NB distribution. The other two models were fitted by the R package of countreg. The full names of cancers are in Supplementary Table S1.