Table 1.
The coefficients of the 74 predictive variables in the three methods
| x | Description | L1LR | BE-AIC | BE-BIC |
|---|---|---|---|---|
| x0 | intercept | 1.09 | 11.47 | 7.63 |
| x1 | largest intensity | 1.48 | - | - |
| x2 | second largest intensity | -1.73 | -4.84 | -4.42 |
| x3 | average of x1 | -1.18 | - | - |
| x4 | average of (x1-x2) | -4.65 | -6.2 | -5.65 |
| x5 | standard error of (x1-x2) | 3.19 | -10.03 | - |
| x6 | 1/x3 | -2.37 | -3.22 | -2.88 |
| x7 | √x5 | 0.54 | 1.42 | 0.77 |
| x8 | log(x5) | -0.93 | 2.69 | - |
| x9 | piecewise function of |x1-x2| | 0.59 | - | - |
| x10 | 3.53 | 4.94 | 4.71 | |
| x11 | 3.45 | 6.62 | 6.3 | |
| x12 | 2.42 | 9.32 | 8.74 | |
| x13 | 1.44 | 12.35 | 11.43 | |
| x14 | 0.34 | 15.41 | 14.21 | |
| x15 | - | 23.06 | 21.45 | |
| x16 | - | 118.87 | 46.79 | |
| x17 | - | - | - | |
| x18 | current cycle number | -0.016 | -0.019 | -0.018 |
| x19 | inverse distance | -0.24 | - | - |
| x20 | indicators of the first 7th cycles | -0.3 | -2.99 | - |
| x21 | -0.15 | - | - | |
| x22 | - | - | - | |
| x23 | - | - | - | |
| x24 | -0.25 | - | - | |
| x25 | -0.54 | -1.22 | - | |
| x26 | 0.32 | 12.49 | - | |
| x27 | A(AC) | -0.11 | - | - |
| x28 | A(AG) | -0.91 | -2.21 | -1.32 |
| x29 | A(AT) | -0.67 | -3.39 | -1.15 |
| x30 | A(CA) | 1.29 | - | - |
| x31 | A(CG) | 0.86 | -2.89 | - |
| x32 | A(CT) | 0.25 | -5.31 | - |
| x33 | A(GA) | 1.44 | -3.23 | - |
| x34 | A(GC) | 0.21 | -5.8 | - |
| x35 | A(GT) | 1.66 | -6.51 | - |
| x36 | A(TA) | 0.89 | -6.96 | - |
| x37 | A(TC) | 0.44 | -8.77 | - |
| x38 | A(TG) | - | -10.79 | - |
| x39 | C(AC) | 2.27 | 2.88 | 2.29 |
| x40 | C(AG) | - | -1.34 | - |
| x41 | C(AT) | - | -2.77 | - |
| x42 | C(CA) | -0.95 | -2.65 | -1.4 |
| x43 | C(CG) | -0.7 | -5.29 | - |
| x44 | C(CT) | -0.7 | -5.29 | - |
| x45 | C(GA) | -1.29 | -7.09 | -1.68 |
| x46 | C(GC) | 0.89 | -3.51 | - |
| x47 | C(GT) | 0.63 | -5.31 | - |
| x48 | C(TA) | 0.68 | -7.14 | - |
| x49 | C(TC) | - | -9.25 | - |
| x50 | C(TG) | -0.54 | -11.32 | - |
| x51 | G(AC) | 0.58 | -1.09 | - |
| x52 | G(AG) | 0.05 | -1.09 | - |
| x53 | G(AT) | -0.45 | -3.32 | -1.1 |
| x54 | G(CA) | 0.18 | -1.4 | - |
| x55 | G(CG) | -0.18 | -4.54 | - |
| x56 | G(CT) | -1.02 | -6.89 | -1.52 |
| x57 | G(GA) | 1.6 | -2.78 | - |
| x58 | G(GC) | 0.24 | -5.76 | - |
| x59 | G(GT) | -0.75 | -9.81 | -1.28 |
| x60 | G(TA) | 0.93 | -7.26 | - |
| x61 | G(TC) | 0.24 | -9.18 | - |
| x62 | G(TG) | 0.7 | -10.12 | - |
| x63 | T(AC) | - | - | - |
| x64 | T(AG) | -0.23 | -1.68 | - |
| x65 | T(AT) | 2.03 | - | - |
| x66 | T(CA) | 0.21 | -1.28 | - |
| x67 | T(CG) | -0.74 | -5.27 | - |
| x68 | T(CT) | -0.1 | -5.72 | - |
| x69 | T(GA) | 0.16 | -4.64 | - |
| x70 | T(GC) | 0.73 | -5.15 | - |
| x71 | T(GT) | 1.94 | -6.55 | - |
| x72 | T(TA) | - | -8.09 | - |
| x73 | T(TC) | -0.29 | -9.76 | - |
| x74 | T(TG) | -0.99 | -11.72 | - |
We denote these 74 variables by x=(x 0,x 1,⋯,x 74). In the first row of the table, ‘L1LR’ means the L 1-regularized logistic regression, ‘BE-AIC’ indicates the backward deletion with AIC, and ‘BE-BIC’ represents the backward deletion with BIC. The details of the variables in each row are described in Methods. x 27 to x 74 are corresponding to the 3-letter sequences, which indicate the type of the base in the previous cycle, type of the base with the largest and the second largest intensity in current cycle. Meanwhile, ‘-’ implies that the method has removed the feature