QSAR – 21st International workshop on QSAR in Environmental and Health Sciences

Andrius Sazonovas^1,2, Kiril Lanevskij^1,2, Alexander Proskura³, Andrey Vazhentsev⁴, Eduard Kolovanov¹

¹ ACD/Labs, Inc., 8 King Street East, Toronto, Ontario, M5C 1B5, Canada
² UAB ACD/Labs Vilnius, J. Savickio g. 4, LT-01108, Vilnius, Lithuania
³ ACD Development Unipessoal Lda, Rua Eng. Ferreira Dias 728, 4100-246, Porto, Portugal
⁴ Advanced Chemistry Development Germany GmbH, Hahnstrasse 70, 60528, Frankfurt am Main, Germany

Compounds with multiple ionization centers exhibit complex behaviors that are important for scientists to understand in many sectors of R&D. For more than 25 years, ACD/Labs has been supporting researchers with prediction algorithms to help with this. The latest improvements in Percepta’s pK_a algorithm directly impact predictions of logD, pH-dependent aqueous solubility, and the distribution of ionic species.

The logD at given pH depends on logP of molecule as well as on percentage of different ionic species of molecule existing at this pH. Ionic species distribution directly depends on all pK_a micro-constants of the molecule.

To calculate the distribution of ionic forms from pH, it is necessary to solve a non-linear system of equations for concentration values where pK_a micro-constants are used as coefficients. Due to the excessive number of pK_a micro-constants, the system of equations turns out to be overprovisioned and therefore has to be solved by the least squares method. Since the micro constants pK_a are predicted by the algorithm with some error, the solution found by the traditional method quite often leads to the situation where the distribution of ionic species, and subsequently the logD curve (or solubility, or any other properties depending on pK_a), deviates significantly from the initially predicted “defining” micro-constants. This raises questions among observers regarding the accuracy of the logD curve prediction itself.

This problem has been resolved in the new version. Now the solution to the overprovisioned system is sought by minimizing the total error of all equations, taking into account the relative importance of a particular micro-constant. For this purpose, instead of the usual Euclidean norm, a weighted norm is used. The procedure for selecting the weighting coefficients of the norm is non-trivial, since the system of equations being solved is nonlinear, and it is also not always obvious which micro-constants should be considered less important or more important.

Andrius Sazonovas^1,2, Kiril Lanevskij^1,2, Alexander Proskura³, Eduard Kolovanov¹

¹ ACD/Labs, Inc., 8 King Street East, Toronto, Ontario, M5C 1B5, Canada
² UAB ACD/Labs Vilnius, J. Savickio g. 4, LT-01108, Vilnius, Lithuania
³ ACD Development Unipessoal Lda, Rua Eng. Ferreira Dias 728, 4100-246, Porto, Portugal

After >25 years of continuous curation of available pK_a data and prediction algorithm development, ACD/Labs, remains committed to excellence in prediction accuracy and model applicability. This effort has earned Percepta’s ionization prediction the reputation of industry leader. Here we present the latest and most substantial improvements to ACD/pK_a resulting in the version 2023 release of Percepta applications which includes, but is not limited to an expansion of the training set by a diverse set of 4000 new compounds obtained via collaborations with our customers from pharma, chemical and environmental industries, introduction of the Hammett-type equations for the new ionization centers not recognized by previous algorithm versions and improvement of the micro pK_a constant prediction speed as well as optimization of the Apparent pK_a selection procedure by prioritizing one preferrable protonation/deprotonation path.

Overall, the implemented changes resulted in a substantial prediction accuracy improvement globally, with an especially notable performance gain for newly added chemical classes, as well as 5-10 times increase in pK_a calculation speed (depending on the dataset), stemming partly from optimization of micro pK_a calculations, but mainly from the new Apparent pK_a selection method. The latter also prevents any possible confusion between acidic and basic ionization center flags for Apparent pK_a values.