Functional group based Ligand binding affinity scoring function at atomic environmental level.

Area of Science:

• Computational chemistry
• Drug discovery
• Structural biology

Background:

• Knowledge-based scoring functions (KBSFs) are crucial for virtual screening and molecular docking in drug discovery.
• Existing KBSFs often rely on distance-based statistical measures and the atom singularity concept, neglecting environmental effects.
• A precise free energy function accounting for all interactions, including water-mediated ones, remains a challenge.

Purpose of the Study:

• To develop a novel knowledge-based statistical energy function for protein-ligand complexes that incorporates atomic environment.
• To address limitations of existing scoring functions by treating functional groups as singular entities.
• To improve the accuracy and efficiency of virtual screening and molecular docking in drug discovery.

Main Methods:

• Developed a novel knowledge-based statistical energy function considering atomic environments and functional groups.
• Designed and utilized a Functional group based Ligand retrieval (FBLR) system to identify and orient functional groups.
• Employed decoy searching with high-resolution protein-ligand complexes to build and validate the KBSF.

Main Results:

• The proposed KBSF effectively accounts for atomic environments, overcoming the atom singularity limitation.
• The FBLR system successfully identifies and orients functional groups within ligands.
• The new KBSF demonstrates speed, simplicity, and improved handling of molecular orientation in active sites.

Conclusions:

• The novel KBSF offers a more accurate and efficient approach to protein-ligand binding prediction.
• The method facilitates de novo drug design through molecular build-up using decoys.
• Potential applications include pharmacophore fragment detection and fragment alignment procedures.