Monte Carlo method for predicting of cardiac toxicity: hERG blocker compounds

Abstract

The estimation of the cardiotoxicity of compounds is an important task for the drug discovery as well as for the risk assessment in ecological aspect. The experimental estimation of the above endpoint is complex and expensive. Hence, the theoretical computational methods are very attractive alternative of the direct experiment. A model for cardiac toxicity of 400 hERG blocker compounds (pIC50) is built up using the Monte Carlo method. Three different splits into the visible training set (in fact, the training set plus the calibration set) and invisible validation sets examined. The predictive potential is very good for all examined splits. The statistical characteristics for the external validation set are (i) the coefficient of determination r(2)=(0.90-0.93); and (ii) root-mean squared error s=(0.30-0.40).

CORAL: Binary classifications (active/inactive) for drug-induced liver injury

“…These are promoters of inactivity of compounds; and (iii) attributes which have in several runs of the Monte Carlo optimization both positive and negative correlation weights, the role of these attributes is not clear. Thus, the suggested approach gives possibility for mechanistic interpretation of a model (Toropov et al, 2012;Gobbi et al, 2016;Veselinović et al, 2016 ). The analysis of three models built up with different distributions into the training, invisible training, calibration, and validation set has shown: there are structural indicators of high probability of liver injury caused by impact of a drug-like substance.…”

“…The CORAL software gives possibility of application of two representations of the molecular structure via SMILES (Gobbi et al, 2016) and via molecular graphs (Toropov et al, 2012). The integrated representation with involving both molecular features extracted from SMILES together with features extracted from graph also is available (Toropov et al, 2012).…”

“…The computational prediction of biochemical endpoints without animal testing is encouraged by various social organizations, for instance, REACH (Registration, Evaluation, Authorisation and Restriction of Chemical Substances) (Pery et al, 2013) and OECD (Organisation for Economic Cooperation and Development) (Gobbi et al, 2016). However it should be noted, the animal testing remains the basis of development of the above computational methods.…”

“…In fact, each therapeutic agent has both positive and negative (or even dangerous) effects (Toropova and Toropov, 2014;Gobbi et al, 2016). Drug-induced liver injury is one of the most common drug-induced impact to human organism leading to life-threatening conditions such as acute liver failure (Persson et al, 2013;Chen et al, 2015;Xu et al, 2015;Zhu and Kruhlak, 2014;Kostadinova et al, 2013), in addition this is one of the leading causes of the termination of drug development projects (Chen et al, 2013).…”

Advances in Computational Prediction of Regioselective and Isoform-Specific Drug Metabolism Catalyzed by CYP450s.

“…QSAR methods are currently one of the most popular ligand based methods for predicting not only metabolic but other forms of toxicity like genotoxicity, skin sensitization, cardiac toxicity, neurotoxicity, and carcinogenicity . In these methods there are a number of pitfalls that one can fall into if proper consideration is not given to a number of points like applicability domain, well defined end points, predictivity, robustness and possibility of mechanistic interpretations etc.…”