QEX: target-specific druglikeness filter enhances ligand-based virtual screening

Area of Science:

• Medicinal Chemistry
• Computational Drug Discovery

Background:

• Druglikeness is crucial for effective drug candidate screening.
• The quantitative estimate of druglikeness (QED) is a widely used metric.
• Existing methods may lack target specificity.

Purpose of the Study:

• To introduce QEX, a novel druglikeness index tailored to individual drug targets.
• To enhance the accuracy and efficiency of drug candidate screening.
• To improve upon the established QED methodology.

Main Methods:

• Developed QEX, a target-specific druglikeness index inspired by QED.
• Modeled physicochemical properties of compounds based on target-specific data.
• Evaluated QEX performance using PubChem assay data and a c-Src inhibitor case study.

Main Results:

• QEX demonstrated superior performance compared to QED in distinguishing active compounds (AUC improvement: 0.069-0.236).
• QEX effectively filtered c-Src inhibitors, outperforming QED.
• QEX proved efficient even with incomplete datasets of inactive compounds.

Conclusions:

• QEX offers improved accuracy and target specificity for druglikeness assessment.
• QEX enhances conventional virtual screening methods, particularly when applied as an initial filter.
• This new index holds significant potential for optimizing drug discovery pipelines.