Metadynamics as a Postprocessing Method for Virtual Screening with Application to the Pseudokinase Domain of JAK2

Area of Science:

• Computational chemistry
• Drug discovery
• Molecular dynamics

Background:

• Standard virtual screening methods often yield false positives, leading to wasted resources on inactive compounds.
• Accurate identification of potential drug candidates is crucial for efficient drug discovery pipelines.

Purpose of the Study:

• To evaluate metadynamics as a postprocessing technique for virtual screening to reduce false positives.
• To assess the ability of metadynamics to differentiate between active and inactive compounds targeting the JAK2 kinase ATP site.

Main Methods:

• Virtual screening using Glide SP to identify potential JAK2 kinase inhibitors.
• Rescoring of top-ranked compounds using Glide XP, GOLD, and MM/GBSA.
• Application of metadynamics to calculate relative binding affinities and ligand dissociation.
• Validation against known binders of varying affinities.

Main Results:

• Standard rescoring methods (Glide XP, GOLD, MM/GBSA) failed to effectively distinguish between active and inactive compounds.
• Metadynamics successfully differentiated between compounds with high, low, or no activity.
• Metadynamics correctly identified most selected virtual screening compounds as false positives.

Conclusions:

• Metadynamics shows significant potential as a postprocessing method to enhance virtual screening accuracy.
• This approach can minimize the financial and temporal costs associated with synthesizing or purchasing inactive compounds.
• Metadynamics offers a viable strategy for improving the reliability of virtual screening campaigns.