We describe recent methodological and parameterization improvements to our ab initio pKa prediction package, Jaguar pKa. This program, which is based on a density functional theory treatment of a thermodynamic cycle augmented with empirical corrections for specific functional groups, can be applied to large drug-like molecules. The introduction of the so-called shell model together with automation of the underlying parameter assignment scheme has allowed us to achieve essentially full coverage of the drug-like regions of chemical space. The spectrum of the high-quality parameterizations for specific functional groups has been greatly extended and the accuracy of pKa predictions generally improved. We present a variety of  results for pharmaceutically relevant test cases obtained with this new version of the program.

Progress in predicting pKa of drug-like molecules with ab initio methods

245th National Meeting (2013)

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.