pharmd

pharmd extracts pharmacophore models from molecular dynamics (MD) trajectories of protein–ligand complexes and uses redundancy reduction, a conformer-coverage ranking approach, and multiple scoring schemes to prioritize compounds for virtual screening in drug discovery.

Key Features:

• Pharmacophore Model Retrieval: Extracts pharmacophore models from MD trajectories of protein–ligand complexes to capture dynamic interactions that static crystal structures may miss.
• Redundancy Reduction: Selects distinct representative pharmacophores by removing those with identical three-dimensional (3D) pharmacophore hashes.
• Conformer Coverage Approach: Ranks compounds by assessing conformer coverage across all representative pharmacophores to account for diverse conformational states.
• Scoring and Ranking: Implements multiple scoring schemes and supports ranking by averaged predicted scores from different complexes.

Scientific Applications:

• Virtual Screening for Drug Discovery: Prioritizes primary hits within large compound libraries during virtual screening workflows.
• Study of Kinase Complexes: Applied to analyze dynamic protein–ligand interactions for targets including cyclin-dependent kinases (CDKs), exemplified by CDK2 complexes.

Methodology:

Extracts pharmacophore models from MD trajectories; removes pharmacophores with identical 3D pharmacophore hashes to select representatives; ranks compounds using a conformer coverage strategy across all representative pharmacophores; applies multiple scoring schemes including averaged predicted scores from different complexes.