Determination of hERG channel blockers using a decision tree

Area of Science:

• Pharmacology
• Computational Chemistry
• Drug Safety

Background:

• Torsade de Pointes (TdP) is a life-threatening arrhythmia linked to QT-interval prolongation.
• QT-interval prolongation is often caused by inhibition of the human Ether-à-go-go-Related Gene (hERG) potassium channel.
• Predicting drug-induced hERG channel blockade is crucial for cardiovascular safety.

Purpose of the Study:

• To develop an in silico method for predicting Torsade de Pointes (TdP)-causing drugs.
• To identify key structural properties of hERG channel blockers.
• To provide guidelines for designing safer drug candidates.

Main Methods:

• Utilized a decision tree approach for drug property prediction.
• Employed molecular modeling and semi-empirical AM1 calculations for hERG channel blockers.
• Derived a pharmacophoric SMARTS string from high-affinity compounds.
• Searched the PubChem database for compounds with QT-prolonging activity.

Main Results:

• The derived SMARTS string was the most significant descriptor in the decision tree.
• Identified novel compounds with QT-prolonging activity in PubChem.
• The model successfully predicted potential TdP-causing drug properties.

Conclusions:

• The decision tree approach effectively predicts TdP-inducing drug potential.
• The pharmacophoric SMARTS string is a valuable tool for drug safety assessment.
• This method aids in the rational design of safer pharmaceutical compounds.