On the prediction of statistical parameters in high-throughput screening using resampling techniques.

Area of Science:

• Drug discovery
• Chemical biology
• Pharmacology

Background:

• High-throughput screening (HTS) is crucial for drug discovery but suffers from false positives and negatives.
• False positives increase project costs, while false negatives (non-hits) represent missed opportunities.
• Accurate prediction of screening rates is essential for efficient HTS campaigns.

Purpose of the Study:

• To investigate methods for predicting primary hit rates, hit confirmation rates, and false-positive/negative rates in HTS.
• To determine optimal compound concentrations and hit thresholds before initiating large-scale screening.
• To enhance the reliability and cost-effectiveness of HTS campaigns.

Main Methods:

• Utilized pilot screen data from approximately 2800 compounds tested at multiple concentrations and replicates.
• Employed a resampling scheme to predict hit and false rates as a function of hit thresholds.
• Used replicate data as internal confirmation to establish confidence limits for predictions.

Main Results:

• Successfully predicted primary hit rate, hit confirmation rate, and false rates based on pilot screen data.
• Demonstrated the accuracy of the prediction method by comparing resampled rates with actual screening campaign results.
• Identified optimal compound concentrations and hit thresholds for minimizing false rates.

Conclusions:

• The developed prediction method accurately forecasts HTS performance metrics.
• Pilot screening with replicates and resampling enables optimization of HTS parameters prior to full-scale campaigns.
• This approach can significantly improve HTS efficiency and reduce associated costs by minimizing false results.