Skip to main content

Table 4 If-Then regression rules with coverage ≥ 10 extracted from the regression tree built by M5P when using as input descriptors both the predicted log K t:p for 12 tissues and a large set of molecular descriptors

From: Predicting volume of distribution with decision tree-based regression methods using predicted tissue:plasma partition coefficients

Rule IF (a set of conditions) THEN Vss (L/Kg) coverage
1 log Kskin:plasma ≤ 0.044 and SlogP_VSA7 ≤ 88.002 0.57 80
2 log Kskin:plasma ≤ 0.044 and SlogP_VSA7 > 88.002 and ASA– ≤ 181.026 0.37 30
3 log Kskin:plasma ≤ 0.044 and SlogP_VSA7 > 88.002 and ASA– > 181.026 0.27 60
4 log Kskin:plasma > 0.044 and a_ICM < = 1.553 and fiA < = 0.017 and PEOE_VSA+0 < = 108.431 and PEOE_RPC– < = 0.187 1.66 13
5 log Kskin:plasma > 0.044 and a_ICM < = 1.553 and fiA < = 0.017 and PEOE_VSA+0 < = 108.431 and PEOE_RPC– > 0.187 1.96 26
6 log Kskin:plasma > 0.044 and a_ICM < = 1.553 and fiA < = 0.017 and PEOE_VSA+0 > 108.431 and SlogP_VSA4 < = 16.939 3.14 20
7 log Kskin:plasma > 0.044 and a_ICM < = 1.553 and fiA < = 0.017 and PEOE_VSA+0 > 108.431 and SlogP_VSA4 > 16.939 2.15 19
8 log Kskin:plasma > 0.044 and a_ICM > 1.553 1.10 100