Christophe Molina

Blockade of the human Ether-à-go-go Related Gene (hERG) potassium channel by small molecules can prolong the QT interval, leading to fatal cardiotoxicity. Numerous drugs have been withdrawn from the market due to cardiac side effects, underscoring the need for early identification of hERG toxicity. Despite several classification machine learning (ML) models having been developed to this end, robustness, class imbalance, and interpretability are still challenges. Using the largest public database of hERG inhibition, this work integrates eXtreme Gradient Boosting (XGBoost) with Isometric Stratified Ensemble (ISE) mapping (XGB + ISE map) to enhance hERG prediction. An XGBoost consensus model was developed using balanced training sets and diverse variable subsets, resulting in robust models less affected by class imbalance. The model demonstrated competitive predictive performance, achieving a balance between sensitivity (SE = 0.83) and specificity (SP = 0.90) through exhaustive validation. ISE mapping estimated the model applicability domain and improved prediction confidence evaluation and compound selection by stratifying data. Refined variable selection procedures enhanced model interpretability. Variable importance analysis highlights key molecular determinants (peoe_VSA8, ESOL, SdssC, MaxssO, nRNR2, MATS1i, nRNHR, nRNH2) associated with hERG inhibition. The XGB + ISE map strategy provides an effective approach to identifying promising molecules in drug discovery campaigns with reduced hERG inhibition risk.

The heterogeneity of the Caco-2 cell line and differences in experimental protocols for permeability assessment using this cell-based method have resulted in the high variability of Caco-2 permeability measurements. These problems have limited the generation of large datasets to develop accurate and applicable regression models. This study presents a QSPR approach developed on the KNIME analytical platform and based on a structurally diverse dataset of over 4900 molecules. Interpretable models were obtained using random forest supervised recursive algorithms for data cleaning and feature selection. The development of a conditional consensus model based on regional and global regression random forest produced models with RMSE values between 0.43–0.51 for all validation sets. The potential applicability of the model as a surrogate for the in vitro Caco-2 assay was demonstrated through blind prediction of 32 drugs recommended by the International Council for the Harmonization of Technical Requirements for Pharmaceuticals (ICH) for validation of in vitro permeability methods. The model was validated for the preliminary estimation of the BCS/BDDCS class. The KNIME workflow developed to automate new drug prediction is freely available. The results suggest that this automated prediction platform is a reliable tool for identifying the most promising compounds with high intestinal permeability during the early stages of drug discovery.

Partial and incremental stratification analysis of a quantitative structure-interference relationship (QSIR) is a novel strategy intended to categorize classification provided by machine learning techniques. It is based on a 2D mapping of classification statistics onto two categorical axes: the degree of consensus and level of applicability domain. An internal cross-validation set allows to determine the statistical performance of the ensemble at every 2D map stratum and hence to define isometric local performance regions with the aim of better hit ranking and selection. During training, isometric stratified ensembles (ISE) applies a recursive decorrelated variable selection and considers the cardinal ratio of classes to balance training sets and thus avoid bias due to possible class imbalance. To exemplify the interest of this strategy, three different highly imbalanced PubChem pairs of AmpC β-lactamase and cruzain inhibition assay campaigns of colloidal aggregators and complementary aggregators data set available at the AGGREGATOR ADVISOR predictor web page were employed. Statistics obtained using this new strategy show outperforming results compared to former published tools, with and without a classical applicability domain. ISE performance on classifying colloidal aggregators shows from a global AUC of 0.82, when the whole test data set is considered, up to a maximum AUC of 0.88, when its highest confidence isometric stratum is retained.

Computational models for predicting aqueous solubility from the molecular structure represent a promising strategy from the perspective of drug design and discovery. Since the first "Solubility Challenge", these initiatives have marked the state-of-art of the modelling algorithms used to predict drug solubility. In this regard, the quality of the input experimental data and its influence on model performance has been frequently discussed. In our previous study, we developed a computational model for aqueous solubility based on recursive random forest approaches. The aim of the current commentary is to analyse the performance of this already trained predictive model on the molecules of the second "Solubility Challenge". Even when our training set has inconsistencies related to the pH, solid form and temperature conditions of the solubility measurements, the model was able to predict the two sets from the second "Solubility Challenge" with statistics comparable to those of the top ranked models. Finally, we provided a KNIME automated workflow to predict aqueous solubility of new drug candidates, during the early stages of drug discovery and development, for ensuring the applicability and reproducibility of our model. (C) 2021 by the authors.

In silico prediction of antileishmanial activity using quantitative structure–activity relationship (QSAR) models has been developed on limited and small datasets. Nowadays, the availability of large and diverse high-throughput screening data provides an opportunity to the scientific community to model this activity from the chemical structure. In this study, we present the first KNIME automated workflow to modeling a large, diverse, and highly imbalanced dataset of compounds with antileishmanial activity. Because the data is strongly biased toward inactive compounds, a novel strategy was implemented based on the selection of different balanced training sets and a further consensus model using single decision trees as the base model and three criteria for output combinations. The decision tree consensus was adopted after comparing its classification performance to consensuses built upon Gaussian-Naı̈ve-Bayes, Support-Vector-Machine, Random-Forest, Gradient-Boost, and Multi-Layer-Perceptron base models. All these consensuses were rigorously validated using internal and external test validation sets and were compared against each other using Friedman and Bonferroni–Dunn statistics. For the retained decision tree-based consensus model, which covers 100% of the chemical space of the dataset and with the lowest consensus level, the overall accuracy statistics for test and external sets were between 71 and 74% and 71 and 76%, respectively, while for a reduced chemical space (21%) and with an incremental consensus level, the accuracy statistics were substantially improved with values for the test and external sets between 86 and 92% and 88 and 92%, respectively. These results highlight the relevance of the consensus model to prioritize a relatively small set of active compounds with high prediction sensitivity using the Incremental Consensus at high level values or to predict as many compounds as possible, lowering the level of Incremental Consensus. Finally, the workflow developed eliminates human bias, improves the procedure reproducibility, and allows other researchers to reproduce our design and use it in their own QSAR problems.

In-silico prediction of aqueous solubility plays an important role during the drug discovery and development processes. For many years, the limited performance of in-silico solubility models has been attributed to the lack of high-quality solubility data for pharmaceutical molecules. However, some studies suggest that the poor accuracy of solubility prediction is not related to the quality of the experimental data and that more precise methodologies (algorithms and/or set of descriptors) are required for predicting aqueous solubility for pharmaceutical molecules. In this study a large and diverse database was generated with aqueous solubility values collected from two public sources; two new recursive machine-learning approaches were developed for data cleaning and variable selection, and a consensus model based on regression and classification algorithms was created. The modeling protocol, which includes the curation of chemical and experimental data, was implemented in KNIME, with the aim of obtaining an automated workflow for the prediction of new databases. Finally, we compared several methods or models available in the literature with our consensus model, showing results comparable or even outperforming previous published models.

In-silico prediction of human oral bioavailability is a relevant tool for the selection of potential drug candidates and for the rejection of those molecules with less probability of success during the early stages of drug discovery and development. However, the high variability and complexity of oral bioavailability and the limited experimental data in public domain have mainly restricted the devel-opment of reliable in-silico models to predict this property from the chemical structure. In this study we present a KNIME automated workflow to predict human oral bioavailability of new drug and drug-like molecules, based on five machine learning approaches combined into an ensemble model. The workflow is freely accessible and allows the quickly and easily prediction of oral bioa-vailability for new molecules, where users do not require any knowledge or advanced experience in machine learning or statistical modeling to automatically obtain their predictions, increasing the potential use of the present proposal.