Open-source QSAR models for pKa prediction using multiple machine learning approaches

Abstract

Background The logarithmic acid dissociation constant pKa reflects the ionization of a chemical, which affects lipophilicity, solubility, protein binding, and ability to pass through the plasma membrane. Thus, pKa affects chemical absorption, distribution, metabolism, excretion, and toxicity properties. Multiple proprietary software packages exist for the prediction of pKa, but to the best of our knowledge no free and open-source programs exist for this purpose. Using a freely available data set and three machine learning approaches, we developed open-source models for pKa prediction. Methods The experimental strongest acidic and strongest basic pKa values in water for 7912 chemicals were obtained from DataWarrior, a freely available software package. Chemical structures were curated and standardized for quantitative structure–activity relationship (QSAR) modeling using KNIME, and a subset comprising 79% of the initial set was used for modeling. To evaluate different approaches to modeling, several datasets were constructed based on different processing of chemical structures with acidic and/or basic pKas. Continuous molecular descriptors, binary fingerprints, and fragment counts were generated using PaDEL, and pKa prediction models were created using three machine learning methods, (1) support vector machines (SVM) combined with k-nearest neighbors (kNN), (2) extreme gradient boosting (XGB) and (3) deep neural networks (DNN). Results The three methods delivered comparable performances on the training and test sets with a root-mean-squared error (RMSE) around 1.5 and a coefficient of determination (R2) around 0.80. Two commercial pKa predictors from ACD/Labs and ChemAxon were used to benchmark the three best models developed in this work, and performance of our models compared favorably to the commercial products. Conclusions This work provides multiple QSAR models to predict the strongest acidic and strongest basic pKas of chemicals, built using publicly available data, and provided as free and open-source software on GitHub.

Current Strategies in Assessment of Nanotoxicity: Alternatives to In Vivo Animal Testing

“…In most cases, 2D-and 3D-QSAR studies are commonly used to evaluate the series of chemical compounds [86][87][88]. The 1D-QSAR approach allows for the determination of correlations for 1D descriptors (pKa, log P, structural fragments, and fingerprints) with biological activity [89]. To date, the 2D-QSAR method has been widely explored in many studies for the study of toxic or medicinal chemistry, which was first proposed in the early 1970s [90].…”

Overview of the SAMPL6 pKaChallenge: Evaluating small molecule microscopic and macroscopic pKapredictions

“…All of the top-performing empirical methods were developed as commercial software that requires a license to run, and there were not any open-source alternatives for empirical p K a predictions. Since the completion of the blind challenge, two publications reported open-source machine learning-based p K a prediction methods, however, one can only predict the most acidic or most basic macroscopic p K a values of a molecule [54] and the second one is only trained for predicting p K a values of monoprotic molecules [55]. Recently, a p K a prediction methodology was published that describes a mixed approach of semi-empirical QM calculations and machine learning that can predict macroscopic p K a s of both mono- and polyprotic species [56].…”

Artificial Intelligence in Drug Design: Algorithms, Applications, Challenges and Ethics

“…In another study, three open-source QSAR models were built to predict the pKa of acids and bases: SVM in conjunction with k-NN, extreme gradient boosting and DNN [186]. PaDEL was used to produce molecular descriptors, fingerprints and fragment counts.…”