Enhancing Carbon Acid pKa Prediction by Augmentation of Sparse Experimental Datasets with Accurate AIBL (QM) Derived Values

The prediction of the aqueous pK(a) of carbon acids by Quantitative Structure Property Relationship or cheminformatics-based methods is a rather arduous problem. Primarily, there are insufficient high-quality experimental data points measured in homogeneous conditions to allow for a good global model to be generated. In our computationally efficient pK(a) prediction method, we generate an atom-type feature vector, called a distance spectrum, from the assigned ionisation atom, and learn coefficients for those atom-types that show the impact each atom-type has on the pK(a) of the ionisable centre. In the current work, we augment our dataset with pK(a) values from a series of high performing local models derived from the Ab Initio Bond Lengths method (AIBL). We find that, in distilling the knowledge available from multiple models into one general model, the prediction error for an external test set is reduced compared to that using literature experimental data alone.

Automated summary

Predicting the aqueous pK(a) of carbon acids is challenging due to insufficient experimental data, but using atom-type feature vectors can help reduce prediction errors. Incorporating knowledge from multiple models into a general model has shown to improve predictions compared to using literature experimental data alone.