Support vector machines classification of hERG liabilities based on atom types.

Drug-induced long QT syndrome (LQTS) can cause critical cardiovascular side effects and has accounted for the withdrawal of several drugs from the market. Blockade of the potassium ion channel encoded by the human ether-a-go-go-related gene (hERG) has been identified as a major contributor to drug-induced LQTS. Experimental measurement of hERG activity for each compound in development is costly and time-consuming, thus it is beneficial to develop a predictive hERG model. Here, we present a hERG classification model formulated using support vector machines (SVM) as machine learning method and using atom types as molecular descriptors. The training set used in this study was composed of 977 corporate compounds with hERG activities measured under the same conditions. The impact of soft margin and kernel function on the performance of the SVM models was examined. The robustness of SVM was evaluated by comparing the predictive power of the models built with 90%, 50%, and 10% of the training set data. The final SVM model was able to correctly classify 94% of an external testing set containing 66 drug molecules. The most important atom types with respect to discriminative power were extracted and analyzed.

[1]  G. Keserü.  Prediction of hERG potassium channel affinity by traditional and hologram qSAR methods. , 2003, Bioorganic & medicinal chemistry letters.

[2]  L. Lang.  High clinical trials attrition rate is boosting drug development costs , 2004 .

[3]  A. Bianucci,et al.  Prediction of hERG potassium channel affinity by the CODESSA approach. , 2006, Bioorganic & medicinal chemistry.

[4]  Tom Fawcett,et al.  An introduction to ROC analysis , 2006, Pattern Recognit. Lett..

[5]  M. Sanguinetti,et al.  A mechanistic link between an inherited and an acquird cardiac arrthytmia: HERG encodes the IKr potassium channel , 1995, Cell.

[6]  Gabriele Cruciani,et al.  Predictive models for hERG potassium channel blockers. , 2005, Bioorganic & medicinal chemistry letters.

[7]  Liang Guo,et al.  Automated electrophysiology in the preclinical evaluation of drugs for potential QT prolongation. , 2005, Journal of pharmacological and toxicological methods.

[8]  Maurizio Recanatini,et al.  Safety of Non-Antiarrhythmic Drugs that Prolong the QT Interval or Induce Torsade de Pointes , 2002, Drug safety.

[9]  Cavero,et al.  QT interval prolongation by non-cardiovascular drugs: issues and solutions for novel drug development. , 1999, Pharmaceutical science & technology today.

[10]  O. Pongs,et al.  Screening lead compounds for QT interval prolongation. , 2001, Drug discovery today.

[11]  Gisbert Schneider,et al.  A Virtual Screening Method for Prediction of the hERG Potassium Channel Liability of Compound Libraries , 2002, Chembiochem : a European journal of chemical biology.

[12]  Harry J Witchel,et al.  The hERG potassium channel as a therapeutic target , 2007, Expert opinion on therapeutic targets.

[13]  Bernard F. Buxton,et al.  Drug Design by Machine Learning: Support Vector Machines for Pharmaceutical Data Analysis , 2001, Comput. Chem..

[14]  Hongmao Sun.  A naive bayes classifier for prediction of multidrug resistance reversal activity on the basis of atom typing. , 2005, Journal of medicinal chemistry.

[15]  G. Robertson,et al.  HERG, a human inward rectifier in the voltage-gated potassium channel family. , 1995, Science.

[16]  Vladimir Vapnik,et al.  Statistical learning theory , 1998 .

[17]  J. Kiehn,et al.  The cardiac hERG/IKr potassium channel as pharmacological target: structure, function, regulation, and clinical applications. , 2006, Current pharmaceutical design.

[18]  Max K Leong,et al.  A novel approach using pharmacophore ensemble/support vector machine (PhE/SVM) for prediction of hERG liability. , 2007, Chemical research in toxicology.

[19]  F. Lombardo,et al.  Experimental and computational approaches to estimate solubility and permeability in drug discovery and development settings , 1997 .

[20]  M. Cadene,et al.  X-ray structure of a voltage-dependent K+ channel , 2003, Nature.

[21]  William Stafford Noble,et al.  Support vector machine , 2013 .

[22]  Y. Z. Chen,et al.  Computer prediction of cardiovascular and hematological agents by statistical learning methods. , 2007, Cardiovascular & hematological agents in medicinal chemistry.

[23]  A. Cavalli,et al.  QT prolongation through hERG K+ channel blockade: Current knowledge and strategies for the early prediction during drug development , 2005, Medicinal research reviews.

[24]  M T D Cronin,et al.  Prediction of hERG K+ blocking potency: Application of structural knowledge , 2004, SAR and QSAR in environmental research.

[25]  Nello Cristianini,et al.  An introduction to Support Vector Machines , 2000 .

[26]  Christine Williams,et al.  Patch clamping by numbers. , 2004, Drug discovery today.

[27]  I. Kola,et al.  Can the pharmaceutical industry reduce attrition rates? , 2004, Nature Reviews Drug Discovery.

[28]  Bernhard Schölkopf,et al.  Kernel Methods in Computational Biology , 2005 .

[29]  B. Ganetzky,et al.  Neurogenetic analysis of potassium currents in Drosophila: synergistic effects on neuromuscular transmission in double mutants. , 1983, Journal of neurogenetics.

[30]  D. Bojanic,et al.  Keynote review: in vitro safety pharmacology profiling: an essential tool for successful drug development. , 2005, Drug discovery today.

[31]  A. Camm,et al.  Relationships between preclinical cardiac electrophysiology, clinical QT interval prolongation and torsade de pointes for a broad range of drugs: evidence for a provisional safety margin in drug development. , 2003, Cardiovascular research.

[32]  Hongmao Sun,et al.  An Accurate and Interpretable Bayesian Classification Model for Prediction of hERG Liability , 2006, ChemMedChem.

[33]  E. Green,et al.  A molecular basis for cardiac arrhythmia: HERG mutations cause long QT syndrome , 1995, Cell.

[34]  T. Nishikawa,et al.  A discriminant model constructed by the support vector machine method for HERG potassium channel inhibitors. , 2005, Bioorganic & medicinal chemistry letters.

[35]  Thorsteinn S. Rögnvaldsson,et al.  Why neural networks should not be used for HIV-1 protease cleavage site prediction , 2004, Bioinform..

[36]  Hongmao Sun,et al.  A Universal Molecular Descriptor System for Prediction of LogP, LogS, LogBB, and Absorption , 2004, J. Chem. Inf. Model..

[37]  D Haussler,et al.  Knowledge-based analysis of microarray gene expression data by using support vector machines. , 2000, Proceedings of the National Academy of Sciences of the United States of America.

[38]  William Stafford Noble,et al.  Support vector machine classification on the web , 2004, Bioinform..

[39]  E. Campbell,et al.  Crystal Structure of a Mammalian Voltage-Dependent Shaker Family K+ Channel , 2005, Science.

[40]  Scott Boyer,et al.  Development, interpretation and temporal evaluation of a global QSAR of hERG electrophysiology screening data , 2007, J. Comput. Aided Mol. Des..

[41]  Youxing Jiang,et al.  Crystal structure and mechanism of a calcium-gated potassium channel , 2002, Nature.

[42]  Alex M Aronov,et al.  Predictive in silico modeling for hERG channel blockers. , 2005, Drug discovery today.

[43]  Matthew Clark,et al.  Development and Evaluation of an in Silico Model for hERG Binding , 2006, J. Chem. Inf. Model..

[44]  G. Vincent.  Prevention of Drug‐Induced Long QT Syndrome: Gender, Chemistry, and Education , 2001, Journal of cardiovascular electrophysiology.

[45]  Vladimir N. Vapnik,et al.  The Nature of Statistical Learning Theory , 2000, Statistics for Engineering and Information Science.

[46]  M. Sanguinetti,et al.  hERG potassium channels and cardiac arrhythmia , 2006, Nature.

[47]  Michael C Hutter,et al.  Determination of hERG channel blockers using a decision tree. , 2006, Bioorganic & medicinal chemistry.