Improvement of virtual screening results by docking data feature analysis.

Area of Science:

• Computational chemistry
• Cheminformatics
• Drug discovery

Background:

• Virtual screening (VS) is crucial for identifying potential drug candidates.
• Evaluating the performance of VS experiments and docking programs is essential.
• Existing methods for VS evaluation can be computationally expensive and require expert intervention.

Purpose of the Study:

• To propose and evaluate a novel approach for assessing virtual screening experiments.
• To introduce Docking Data Feature Analysis (DDFA) as an efficient VS evaluation tool.
• To demonstrate the adaptability of DDFA across multiple docking programs.

Main Methods:

• Developed Docking Data Feature Analysis (DDFA) involving two steps.
• Computed features from docking output data for each molecule.
• Utilized an artificial neural network (ANN) to estimate molecular activity based on computed features.

Main Results:

• DDFA achieved outstanding results on the Directory of Useful Decoys (DUD) benchmark.
• Performance surpassed conventional docking program rankings and literature methods.
• DDFA demonstrated comparable results to state-of-the-art methods with reduced resource requirements.

Conclusions:

• DDFA is an automatic and highly attractive methodology for virtual screening evaluation.
• The approach offers a computationally efficient and adaptable solution for VS.
• DDFA provides a valuable tool for accelerating drug discovery processes.