In Silico Prediction of Skin Permeability Using a Two-QSAR Approach

Table S1. Selected compounds for this study; their names, SMILES strings, CAS numbers, and observed log Kp values; their predicted values by SVR A, SVR B, SVR C, HSVR, PLS, and CDC SPC; data partitions; and references; Table S2. Optimal runtime parameters for the SVR models; Table S3. Compound source for the mock test, their names, IUPAC names, CAS numbers, SMILES strings, observed log Pe values, observed log Kp values, and predicted values by SVR A, SVR B, SVR C, and HSVR. Figure S1. Histogram representation of the distributions of various descriptors for all molecules in the training set, test set, and outlier set. (A) log Kp, (B) molecular weight (MW), (C) molecular volume (Vm), (D) n-octanol-water partition coefficient (log P), (E) number of hydrogen bond acceptor (HBA), and (F) number of hydrogen bond donor (HBD) in the training set, test set, and outlier set