HergSPred: Accurate Classification of hERG Blockers/Nonblockers with Machine-Learning Models

Abstract

The human ether-à-go-go-related gene (hERG) K+ channel plays an important role in cardiac action potentials. The inhibition of the hERG channel may lead to long QT syndrome (LQTS) and even sudden cardiac death. Due to severe hERG-related cardiotoxicity, many drugs have been withdrawn from the market. Therefore, it is necessary to estimate the chemical blockade of hERG in the early stage of drug discovery. In this study, we collected 12,850 compounds with hERG inhibition data from the literature and trained a series of hERG blocking classification models based on the MACCS and Morgan fingerprints. A consensus model named HergSPred was generated based on the individual models using voting principles. The accuracy of HergSPred is higher than previous models using identical training and test sets. Moreover, we analyzed the contribution of each input fingerprint to the prediction output to obtain intuitive chemical insights into the hERG inhibition, which allows visualization of warning substructures that may cause cardiotoxicity in the input compound. The model is available at .

Discovery of Orally Bioavailable SOS1 Inhibitors for Suppressing KRAS-Driven Carcinoma

“…Recently, many drugs have been withdrawn from the market due to sudden cardiac death . In the preclinical study stage, 24% of the drugs were discontinued due to cardiovascular side effects, and 45% of the drugs were withdrawn from the market due to cardiac side effects .…”

“…Recently, many drugs have been withdrawn from the market due to sudden cardiac death. 27 In the preclinical study stage, 24% of the drugs were discontinued due to cardiovascular side effects, and 45% of the drugs were withdrawn from the market due to cardiac side effects. 28 The leading causes of cardiac toxicity caused by drugs are blocking the fast delayed rectifier current (IKr) of the heart, prolonging the QT interval in the duration of a cardiac action potential and then inducing torsade de pointe, which can cause sudden death in severe cases.…”

Biological Activity Predictions of Ligands Based on Hybrid Molecular Fingerprinting and Ensemble Learning

“…17 average error of the validation set, selects a model with higher determination coefficient to predict the test sample, and adds the average error to the final prediction value. 18,19 The reason for the integration of the BPC model and SVR model is that the machine learning model has different effects on different data sets, that is, the model has unstable prediction effects on multiple different data sets. The integration of two different models can improve the overall stability of the model.…”

“…Different validation sets correspond to different training sets, and different models can be obtained by using BP models with the same parameters. Each BP network calculates the determination coefficient and average error of the validation set, selects a model with higher determination coefficient to predict the test sample, and adds the average error to the final prediction value. , …”

“…Each BP network calculates the determination coefficient and average error of the validation set, selects a model with higher determination coefficient to predict the test sample, and adds the average error to the final prediction value. 18 , 19 …”