Screening Explorer-An Interactive Tool for the Analysis of Screening Results

Area of Science:

• Computational chemistry
• Bioinformatics
• Drug discovery

Background:

• Traditional analysis of screening results involves separate evaluation of ROC, predictiveness, and enrichment curves and their metrics.
• Calculating partial metrics for different data fractions is repetitive and lacks integrated visualization.
• Existing tools do not offer simultaneous viewing of partial metrics across different charts.

Purpose of the Study:

• To develop an intuitive web-based application, Screening Explorer, for comprehensive evaluation of screening experiment results.
• To integrate complementary metrics (ROC, predictiveness, enrichment) interactively for a deeper understanding.
• To implement consensus scoring methods for evaluating combined enrichment strategies.

Main Methods:

• Developed a web-based application, Screening Explorer.
• Integrated interactive visualization of ROC, predictiveness, and enrichment curves and their metrics.
• Implemented consensus scoring based on score normalization, standardization (z-scores), and compound ranking.

Main Results:

• Screening Explorer provides an intuitive platform for evaluating screening results by linking complementary metrics.
• The tool enables simultaneous viewing and comparison of partial metrics across different charts.
• Consensus scoring methods were implemented to assess combined enrichment strategies.

Conclusions:

• Screening Explorer simplifies the analysis of virtual and experimental screening results.
• The application facilitates a deeper understanding of screening data through interactive metric linkage.
• This tool enhances the evaluation process in drug discovery and development.