1. Abbreviations of solvents: H2O: Water; DMSO: Dimethyl sulfoxide; EtOH_50%: Ethanol:Water=50:50; AN: Acetonitrile; MeOH: Methanol.
2. Prediction methods: 1) XGBoost with RMSE=1.79 and r2=0.918 (80:20 train test split); 2) Neural Network with RMSE=1.60 and and r2=0.930 (80:20 train test split).
3. Experimental data: experimental data comes from the sub-database of iBonD, for more details please click http://ibond.nankai.edu.cn.
4. Note: for special molecule's pKa which is out of the solvent leveling range, the experimental value is unreliable, so we exclude these values from training set. For these molecules, this model will return unreliable results.