Skip to main content

Table 1 PCM, Family QSAR and Family QSAM performance on the PCM dataset

From: Proteochemometric modelling coupled to in silico target prediction: an integrated approach for the simultaneous prediction of polypharmacology and binding affinity/potency of small molecules

  R 2 CV RMSE CV R 2 0 ext RMSE ext
GBM PCM 0.75 0.64 0.79 0.59
GP PCM 0.75 0.65 0.76 0.63
RF PCM 0.74 0.66 0.77 0.62
SVM PCM 0.76 0.63 0.77 0.62
Family QSAM 0.07 1.24 0.09 1.22
Family QSAR 0.61 0.80 0.63 0.78
Inductive Transfer 0.72 0.68 0.76 0.63
  1. Abbreviations: QSAM Quantitative Structure-Activity Modelling, QSAR Quantitative Structure-Activity Relationship, GBM Gradient Boosting Machine, GP Gaussian Process, RF Random Forest, SVM Support Vector Machine.
  2. PCM, with R2 0 test and RMSEtest values of 0.79 and 0.59 pIC50 units, outperforms both Family QSAR, with R2 0 test and RMSEtest values of 0.63 and 0.78 pIC50 units, respectively, and Family QSAM, with with R2 0 test and RMSEtest values of 0.09 and 1.22 pIC50 units, respectively.