Description:
Regression model to predict inhibitors of angiotensin converting enzyme (ACE). The model was built with the Multiple Linear Regression technique by using a total of 6 QuBiLS-MAS descriptors.
Training and testing datasets:
A total of 76 training compounds and 38 testing compounds were extracted from the Sutherland, et al., 10.1021/jm0497141
Internal performance:
For a 10-fold cross-validation repeated 100 times: Squared R = 0.8200, MAE = 0.7614, and RMSE = 0.9955.
External performance:
Squared R = 0.7880, MAE = 0.8914, and RMSE = 1.1747.
Regression equation:
pLC50 =
0.2152 * AC[6]_S_Q_BB_nCi_2_NS14_H_D_NSRW_m_MAS + -29.3484 * AC[2]_N1_B_AB_nCi_2_SS15_M_LGP[1]_p-est_MAS + 0.6737 * TS[7]_S_Q_AB_nCi_2_SS13_H_X_LGP[1-8]_r_MAS + 2401.6263 * GV[7]_P3_B_AB_nCi_2_MP13_n_C_LGP[4-8]_p-dc3_MAS + 0.4073 * AC[4]_K_F_BB_nCi_2_NS12_C_LGP[3-7]_ec_MAS + 0.0137 * ES_N1_Q_AB_nCi_2_SS1_n_T_LGP[1-6]_dc3_MAS + 1.8127