Novel Bayesian classification models for predicting compounds blocking hERG potassium channels

Li, Lili; Lü, Jing; Yin, Ling; Zheng, Mingyue; Luo, Xiaomin; Zhu, Weiliang; Jiang, Hualiang; Chen, Kaixian

doi:10.1038/aps.2014.35

Cited by 51 publications

(28 citation statements)

References 56 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Liu et al . developed the Bayesian classification model using four molecular properties (MW, PSA, AlogP, and pKa_basic), as well as extended-connectivity fingerprints (ECFP_4), based on a dataset of 2,644 compounds including compounds tested on hERG in the literature and FDA-approved drugs, divided into a training set of 2,389 compounds and a test set of 255 compounds 19 . In addition, further validation was performed experimentally using an external data set of 60 compounds by Doddareddy 20 .…”

Section: Introductionmentioning

confidence: 99%

Support Vector Machine model for hERG inhibitory activities based on the integrated hERG database using descriptor selection by NSGA-II

Ogura¹,

Sato²,

Yuki³

et al. 2019

Sci Rep

View full text Add to dashboard Cite

Assessing the hERG liability in the early stages of drug discovery programs is important. The recent increase of hERG-related information in public databases enabled various successful applications of machine learning techniques to predict hERG inhibition. However, most of these researches constructed the datasets from only one database, limiting the predictability and scope of the models. In this study, a hERG classification model was constructed using the largest dataset for hERG inhibition built by integrating multiple databases. The integrated dataset consisted of more than 291,000 structurally diverse compounds derived from ChEMBL, GOSTAR, PubChem, and hERGCentral. The prediction model was built by support vector machine (SVM) with descriptor selection based on Non-dominated Sorting Genetic Algorithm-II (NSGA-II) to optimize the descriptor set for maximum prediction performance with the minimal number of descriptors. The SVM classification model using 72 selected descriptors and ECFP_4 structural fingerprints recorded kappa statistics of 0.733 and accuracy of 0.984 for the test set, substantially outperforming the prediction performance of the current commercial applications for hERG prediction. Finally, the applicability domain of the prediction model was assessed based on the molecular similarity between the training set and test set compounds.

show abstract

Section: Introductionmentioning

confidence: 99%

Support Vector Machine model for hERG inhibitory activities based on the integrated hERG database using descriptor selection by NSGA-II

Ogura¹,

Sato²,

Yuki³

et al. 2019

Sci Rep

View full text Add to dashboard Cite

show abstract

“…This permits applications of automated electrophysiology in primary screening. In addition to experimental approaches, to predict the hERG liability of lead candidates early in drug discovery, numerical in silico methods have been developed [12][13][14][15][16][17][18] . Since experimental data from a large collection of structurally diverse compounds were yet available, models was usually based on the existing data of limited number of hERG inhibitors generated using different methods.…”

mentioning

confidence: 99%

Investigation of miscellaneous hERG inhibition in large diverse compound collection using automated patch-clamp assay

Zou

Wang

et al. 2016

Acta Pharmacol Sin

View full text Add to dashboard Cite

Aim: hERG potassium channels display miscellaneous interactions with diverse chemical scaffolds. In this study we assessed the hERG inhibition in a large compound library of diverse chemical entities and provided data for better understanding of the mechanisms underlying promiscuity of hERG inhibition.Methods: Approximately 300 000 compounds contained in Molecular Library Small Molecular Repository (MLSMR) library were tested. Compound profiling was conducted on hERG-CHO cells using the automated patch-clamp platform-IonWorks Quattro TM .Results: The compound library was tested at 1 and 10 μmol/L. IC 50 values were predicted using a modified 4-parameter logistic model. Inhibitor hits were binned into three groups based on their potency: high (IC 50 <1 μmol/L), intermediate (1 μmol/L< IC 50 <10 μmol/L), and low (IC 50 >10 μmol/L) with hit rates of 1.64%, 9.17% and 16.63%, respectively. Six physiochemical properties of each compound were acquired and calculated using ACD software to evaluate the correlation between hERG inhibition and the properties: hERG inhibition was positively correlative to the physiochemical properties ALogP, molecular weight and RTB, and negatively correlative to TPSA. Conclusion: Based on a large diverse compound collection, this study provides experimental evidence to understand the promiscuity of hERG inhibition. This study further demonstrates that hERG liability compounds tend to be more hydrophobic, high-molecular, flexible and polarizable.Keywords: hERG; high-throughput screening; MLSMR library; automated electrophysiology; cardiotoxicity Acta Pharmacologica Sinica (2016) 37: 111−123; doi: 10.1038/aps.2015 Original Article IntroductionThe acquired long QT syndrome is both a threat to public health and a major stumbling block for drug development [1] . It is most often caused through unintended blockade of the cardiac repolarizing potassium channel, I Kr , encoded by the Human Ether-a-go-go related gene (hERG) [2] . Blockade of hERG channel was found to be associated with an increased duration of ventricular repolarization and prolongation of QT interval (long QT syndrome, or LQTS). hERG potassium channel displays promiscuous interactions with diverse chemical scaffolds [3] . Structurally and functionally unrelated drugs have been shown to block hERG channel, and some of these agents, including terfenadine, astemizole, droperidol and cisapride, etc, have been withdrawn from the market due to their potential to predispose individuals to sudden cardiac death. Due to the important implications in both drug discovery [4] and clinical practice, evaluation of hERG liability has been required for any new chemical entities by regulatory agencies according to the ICH S7B guideline since 2005 [5] .Manual patch clamp method is considered as gold standard for ion channel functional evaluation [6] . However the method is very labor-intensive with low throughput, and thus its application in drug discovery is limited. Ion flux assays with surrogate ions of Rubidium (Rb + ) or Thallium (Tl + ) have ...

show abstract

“…These authors built two models using NB classification and recursive partitioning techniques and found that the NB model yielded better prediction results. Recently, Liu et al (2014) reported a model that was built by laplacsian-corrected Bayesian classification. Molecular fingerprints (extended-connectivity fingerprints) were used as descriptors, and the established models could identify the substructures as favourable or unfavourable for hERG channel blockage.…”

Section: Herg Toxicitymentioning

confidence: 99%

In silicoADME/T modelling for rational drug design

Wang

Xing

Yue

et al. 2015

Quart. Rev. Biophys.

265

155

View full text Add to dashboard Cite

Abstract. In recent decades, in silico absorption, distribution, metabolism, excretion (ADME), and toxicity (T) modelling as a tool for rational drug design has received considerable attention from pharmaceutical scientists, and various ADME/T-related prediction models have been reported. The high-throughput and low-cost nature of these models permits a more streamlined drug development process in which the identification of hits or their structural optimization can be guided based on a parallel investigation of bioavailability and safety, along with activity. However, the effectiveness of these tools is highly dependent on their capacity to cope with needs at different stages, e.g. their use in candidate selection has been limited due to their lack of the required predictability. For some events or endpoints involving more complex mechanisms, the current in silico approaches still need further improvement. In this review, we will briefly introduce the development of in silico models for some physicochemical parameters, ADME properties and toxicity evaluation, with an emphasis on the modelling approaches thereof, their application in drug discovery, and the potential merits or deficiencies of these models. Finally, the outlook for future ADME/T modelling based on big data analysis and systems sciences will be discussed.

show abstract

Novel Bayesian classification models for predicting compounds blocking hERG potassium channels

Cited by 51 publications

References 56 publications

Support Vector Machine model for hERG inhibitory activities based on the integrated hERG database using descriptor selection by NSGA-II

Support Vector Machine model for hERG inhibitory activities based on the integrated hERG database using descriptor selection by NSGA-II

Investigation of miscellaneous hERG inhibition in large diverse compound collection using automated patch-clamp assay

In silicoADME/T modelling for rational drug design

Contact Info

Product

Resources

About