A Machine Learning Algorithm Suggests Repurposing Opportunities for Targeting Selected GPCRs.

Area of Science:

• Computational chemistry
• Pharmacology
• Drug discovery

Background:

• Drug repurposing accelerates the discovery of new therapeutics by utilizing drugs with established safety profiles.
• Computational methods enhance drug repurposing by providing a rational framework and reducing failure rates.

Purpose of the Study:

• To leverage computational approaches for identifying novel drug repurposing opportunities.
• To screen the DrugBank database for potential agonists and antagonists of G-protein coupled receptors (GPCRs).

Main Methods:

• Utilized the Iterative Stochastic Elimination (ISE) algorithm, a machine learning tool, to build ligand-based models.
• Developed agonist and antagonist models for six GPCR families using physicochemical properties and published bioactivity data (IC50, Ki, EC50).
• Screened the DrugBank database against these models to identify top-scoring drug candidates.

Main Results:

• Identified 5982 potential new molecular actions (agonists and antagonists).
• Generated a list of drug repurposing candidates targeting cannabinoid 2 (CB2), histamine (H1, H3, H4), and dopamine 3 (D3) receptors.
• These candidates show potential for treating conditions including neuroinflammation, obesity, allergic dermatitis, and drug abuse.

Conclusions:

• The computational screening successfully identified numerous drug repurposing candidates.
• The identified candidates warrant experimental validation for novel therapeutic applications.
• This approach significantly increases the probability of discovering highly bioactive molecules for unmet medical needs.