Docking structurally similar analogues: Dealing with the false-positive.

Area of Science:

• Medicinal Chemistry
• Computational Drug Discovery
• Pharmacology

Background:

• Analogue design is fundamental to new drug discovery, often employing a fast-follow approach to identify best-in-class therapeutics.
• Virtual screening algorithms like molecular docking are widely used but can yield false positives, particularly for close structural analogues.

Purpose of the Study:

• To investigate the reliability of docking predictions for near-pharmacophoric analogues of the JAK3 inhibitor Tasocitinib.
• To identify the reasons behind false positive docking predictions in drug discovery.
• To propose improved computational strategies for evaluating drug analogues.

Main Methods:

• Molecular docking simulations were performed on two analogues of Tasocitinib.
• Post-docking analysis included MM/GBSA calculations and electrostatic potential mapping.
• Molecular dynamics (MD) simulations were employed to assess complex stability and ligand-protein interactions.

Main Results:

• Two near-pharmacophoric analogues of Tasocitinib yielded positive docking predictions but were experimentally inactive.
• MM/GBSA and electrostatic potential analysis indicated a weakened crucial hydrogen bond in the JAK3 ATP-binding pocket.
• MD simulations revealed instability in the analogue-protein complexes, with ligands detaching within 1.2 ns.

Conclusions:

• Standard docking may produce false positives for 'me-too' drug analogues.
• Integrating post-processing methods like MM/GBSA and MD simulations enhances the predictive accuracy of virtual screening.
• Short post-MD simulations can significantly improve docking predictions for closely related analogues.