Predicting new molecular targets for known drugs

Area of Science:

• Pharmacology
• Computational Chemistry
• Drug Discovery

Background:

• Drugs often interact with multiple physiological targets, leading to both intended effects and side effects.
• Exploring these drug-target interactions is crucial for understanding drug efficacy and safety.

Purpose of the Study:

• To computationally predict novel drug-target associations for a large set of approved and investigational drugs.
• To experimentally validate predicted drug-target interactions and assess their physiological relevance.

Main Methods:

• A chemical similarity approach was used to compare 3,665 US Food and Drug Administration (FDA)-approved and investigational drugs against hundreds of targets defined by their ligands.
• Thirty predicted drug-target associations were selected for experimental testing.

Main Results:

• Thousands of unanticipated drug-target associations were predicted based on chemical similarity.
• Twenty-three new drug-target associations were experimentally confirmed, with five showing high potency (<100 nM).
• The physiological relevance of N,N-dimethyltryptamine (DMT) on serotonergic receptors was confirmed in a knockout mouse model.

Conclusions:

• The chemical similarity approach is a systematic and comprehensive method for predicting drug-target interactions.
• This approach can identify potential drug side effects and suggest new therapeutic indications for existing drugs.