| Abstract
| - The binding affinities to human serum albumin for 94 diverse drugs and drug-like compounds were modeledwith the descriptors calculated from the molecular structure alone using a quantitative structure−activityrelationship (QSAR) technique. The heuristic method (HM) and support vector machine (SVM) were utilizedto construct the linear and nonlinear prediction models, leading to a good correlation coefficient (R2) of0.86 and 0.94 and root-mean-square errors (rms) of 0.212 and 0.134 albumin drug binding affinity units,respectively. Furthermore, the models were evaluated by a 10 compound external test set, yielding R2 of0.71 and 0.89 and rms error of 0.430 and 0.222. The specific information described by the heuristic linearmodel could give some insights into the factors that are likely to govern the binding affinity of the compoundsand be used as an aid to the drug design process; however, the prediction results of the nonlinear SVMmodel seem to be better than that of the HM.
|
