Evaluation of a novel virtual screening strategy using receptor decoy binding sites

Area of Science:

• Computational chemistry
• Biomedical research
• Drug discovery

Background:

• Virtual screening is crucial for identifying potential drug candidates by predicting molecule-protein interactions.
• Accurate ranking of screened molecules is essential for efficient drug discovery pipelines.

Purpose of the Study:

• To develop and evaluate a straightforward receptor-decoy strategy to enhance ranking in receptor-based molecular docking.
• To assess the effectiveness of this strategy in improving ligand prediction accuracy.

Main Methods:

• A receptor-decoy strategy was implemented, involving the definition of a decoy binding site on the receptor.
• The ranking of the true binding-site virtual screen was adjusted based on a decoy-site screen using Autodock Vina.
• Receiver Operator Characteristic Enrichment (ROCE) was used to evaluate performance.

Main Results:

• Improved ROCE was achieved for 5 out of 15 investigated receptor targets when considering up to 15% of the decoy site rank list.
• No improvement in enrichment was observed for 7 targets.
• ROCE was reduced for 3 of the investigated targets.

Conclusions:

• The receptor-decoy strategy shows potential for improving virtual screening enrichment but its effectiveness is highly dependent on the specific target receptor.
• Further research is needed to understand the factors influencing the success of this strategy across different protein targets.