Estimation of coefficients of the logistic regression model used to analyze the transcript levels of ATG genes between IM responders and IM nonresponders
| Variable . | Coefficient estimate . | SD . | P value . |
|---|---|---|---|
| Intercept | 0.891 | 0.317 | <.01* |
| ATG4B | −0.190 | 0.089 | .043* |
| ATG5 | 0.106 | 0.053 | .054 |
| MAP1LC3B | −0.510 | 0.283 | .084 |
| Variable . | Coefficient estimate . | SD . | P value . |
|---|---|---|---|
| Intercept | 0.891 | 0.317 | <.01* |
| ATG4B | −0.190 | 0.089 | .043* |
| ATG5 | 0.106 | 0.053 | .054 |
| MAP1LC3B | −0.510 | 0.283 | .084 |
A logistic model was used to study any possible correlation between ATG genes with significant P values between IM responders and IM nonresponders. The model applied was β0+ β1ATG4B + β2ATG5 + β3MAP1LC3B.
The estimated coefficients and P values of the final model show that the transcript levels of ATG4B have a significant impact at the 5% level.