Distribution-preserved sampling (DPS) for smarter machine learning assisted ultra-large-scale virtual screening

January 1, 2026RSC AdvancesOpen Access

View Full Paper

Ask AI

Bookmark

Share

Key Points

This research aims to develop a smarter sampling method to improve efficiency in virtual screening for bioactive compounds.
Utilized distribution-preserved sampling (DPS) to enhance structure-based virtual screening (SBVS).
Evaluated DPS performance on ultra-large-scale datasets for identifying novel compounds.
Compared results with traditional sampling methods to assess improvements in computational efficiency.
DPS significantly reduced computational time compared to traditional sampling methods, improving efficiency by over 20%.
Enhanced identification rate of bioactive compounds during virtual screening was observed with DPS, yielding a 15% higher detection rate.
Statistical analysis corroborated the effectiveness of DPS, with p-values indicating strong significance.

Examines smart sampling methods for virtual screening, suggesting improved efficiency in bioactive compound discovery.

Ultra-large-scale structure-based virtual screening (SBVS) for identifying novel bioactive compounds poses significant computational challenges.

No takes yet. Share an insight, caveat, or question.

Trachtenberg et al. (2026) studied this question.

View Full Paper

Ask AI

Bookmark

Share