Prediction of hERG Liability – Using SVM Classification, Bootstrapping and Jackknifing

Abstract

Drug-induced QT prolongation leads to life-threatening cardiotoxicity, mostly through blockage of the human ether-a-go-go-related gene (hERG) encoded potassium ion (K+) channels. The hERG channel is one of the most important antitargets to be addressed in the early stage of drug discovery process, in order to avoid more costly failures in the development phase. Using a thallium flux assay, 4,323 molecules were screened for hERG channel inhibition in a quantitative high throughput screening (qHTS) format. Here, we present support vector classification (SVC) models of hERG channel inhibition with the averaged area under the receiver operator characteristics curve (AUC-ROC) of 0.93 for the tested compounds. Both Jackknifing and bootstrapping have been employed to rebalance the heavily biased training datasets, and the impact of these two under-sampling rebalance methods on the performance of the predictive models is discussed. Our results indicated that the rebalancing techniques did not enhance the predictive power of the resulting models; instead, adoption of optimal cutoffs could restore the desirable balance of sensitivity and specificity of the binary classifiers. In an external validation set of 66 drug molecules, the SVC model exhibited an AUC-ROC of 0.86, further demonstrating the utility of this modeling approach to predict hERG liabilities.

Molecular Docking Guided Grid-Independent Descriptor Analysis to Probe the Impact of Water Molecules on Conformational Changes of hERG Inhibitors in Drug Trapping Phenomenon

“…Previously reported MD studies showed that water is mainly present outside the membrane, with lower amounts present within the permeation pore to facilitate ion transportation through the pore [45]. Our findings are in line with previous reports that water is not abundantly present in the pore domain of open channel [44,46,47].…”

Machine learning models for classification tasks related to drug safety

“…Here, we have collected 15 works from the past five years that employ machine learning-based classification approaches to predict hERG inhibition [38][39][40][41][42][43][44][45][46][47][48][49][50][51]. All of these works apply training datasets of more than 1,000 molecules (and up to tens of thousands in some cases [47,48]), and an overall majority presents two-class (active vs. inactive) classification (with the notable example of the 2015 study of Braga et al, who have introduced a third class of "weak blockers") [38].…”

Capsule Networks Showed Excellent Performance in the Classification of hERG Blockers/Nonblockers

“…Furthermore, we used Doddareddy's experimentally validated dataset (a total of 60 compounds: 50 agents from the Chembridge database and 10 from an in-house compound library) as an external validation set to assess the generalization ability of our models. In order to compare the performance of our models with others reported in the literature, we also used the same data sets as those in the literature, including Hou's (Wang et al, 2012;Wang et al, 2016), Zhang's (Zhang et al, 2016), Sun's (Sun et al, 2017), Siramshetty's (Siramshetty et al, 2018), and Cai's (Cai et al, 2019) data sets. Here, it is necessary to mention that an integrated data set of hERG blockade, which is the largest database to date, has been collected by Sato et al (2018).…”