Article

Computer prediction of cardiovascular and hematological agents by statistical learning methods.

Bioinformatics and Drug Design Group, Department of Pharmacy, National University of Singapore, Singapore 117543.
Cardiovascular & Hematological Agents in Medicinal Chemistry (Formerly Current Medicinal Chemistry - Cardiovascular & Hematological Agents) 02/2007; 5(1):11-9. DOI: 10.2174/187152507779315787
Source: PubMed

ABSTRACT Computational methods have been explored for predicting agents that produce therapeutic or adverse effects in cardiovascular and hematological systems. The quantitative structure-activity relationship (QSAR) method is the first statistical learning methods successfully used for predicting various classes of cardiovascular and hematological agents. In recent years, more sophisticated statistical learning methods have been explored for predicting cardiovascular and hematological agents particularly those of diverse structures that might not be straightforwardly modelled by single QSAR models. These methods include partial least squares, multiple linear regressions, linear discriminant analysis, k-nearest neighbour, artificial neural networks and support vector machines. Their application potential has been exhibited in the prediction of various classes of cardiovascular and hematological agents including 1, 4-dihydropyridine calcium channel antagonists, angiotensin converting enzyme inhibitors, thrombin inhibitors, AchE inhibitors, HERG potassium channel inhibitors and blockers, potassium channel openers, platelet aggregation inhibitors, protein kinase inhibitors, dopamine antagonists and torsade de pointes causing agents. This article reviews the strategies, current progresses and problems in using statistical learning methods for predicting cardiovascular and hematological agents. It also evaluates algorithms for properly representing and extracting the structural and physicochemical properties of compounds relevant to the prediction of cardiovascular and hematological agents.

0 Bookmarks
 · 
96 Views
  • [Show abstract] [Hide abstract]
    ABSTRACT: Acquired long QT syndrome causes severe cardiac side effects and represents a major problem in clinical studies of drug candidates. One of the reasons for development of arrhythmias related to long QT is inhibition of the human ether-a-go-go-related-gene (hERG) potassium channel. Therefore, early prediction of hERG K(+) channel affinity of drug candidates is becoming increasingly important in the drug discovery process. Binary QSAR models with threshold values at IC(50)=1 and of 10 microM, respectively, were generated using two different sets of descriptors. One set comprising 32 P_VSA descriptors and the other one utilizing a set of descriptors identified out of a large set via a feature selection algorithm. For the full dataset, the power for classification of hERG blockers was 82-88%, which meets prior classification models. Considering the fact that 2D descriptors are fast and easy to calculate, these binary QSAR models are versatile tools for use in virtual screening protocols.
    Bioorganic & medicinal chemistry 05/2008; 16(7):4107-19. · 2.82 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: 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.
    Bioorganic & medicinal chemistry 07/2008; 16(11):6252-60. · 2.82 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Acquired long QT syndrome caused by drugs that block the human ether-a-go-go-related-gene (hERG) K(+) channel causes severe side effects and thus represents a major problem in clinical studies of drug candidates. Therefore, early prediction of hERG K(+) channel affinity of drug candidates is becoming increasingly important in the drug discovery process. Both structure-based and ligand-based approaches have been undertaken to shed more light on the molecular basis of drug-channel interaction. In this article, in silico approaches for prediction of interaction with hERG are reviewed. Special attention is drawn to the in vitro biological testing systems as well as to consensus approaches for improvement of predictive power.
    Current Medicinal Chemistry 02/2007; 14(28):3003-26. · 3.72 Impact Factor

Full-text

View
2 Downloads
Available from

Chun Wei Yap