Combinatorial Design of Molecule using Activity-Linked Substructural Topological Information as Applied to...

Area of Science:

• Medicinal Chemistry
• Computational Chemistry
• Drug Discovery

Background:

• Combinatorial chemistry accelerates the discovery of bioactive compounds.
• Existing methods lack the ability to generate structures from activity-linked substructural data.

Purpose of the Study:

• Develop a graph-theoretical method for combinatorial structure generation of antitubercular compounds.
• Utilize activity-linked substructural topological information.
• Predict compound activity and prioritize drug candidates.

Main Methods:

• Identify activity-related vertices from compound datasets.
• Generate structures combinatorially using topological distance distributions.
• Predict biological activity using topological indices and a rule-based system.
• Prioritize generated structures with a novel Molecular Priority Score (MPS).

Main Results:

• Successfully generated active antitubercular compounds from substructural information of existing active compounds.
• Achieved a high success rate in activity prediction.
• Identified highly active Acid Alkyl Ester (AAE) compounds with high MPS scores.
• Demonstrated the potential of MPS for prioritizing drug candidates.

Conclusions:

• The developed method shows promise for discovering novel therapeutic candidates for tuberculosis.
• The approach may be applicable to discovering drug molecules for other diseases.
• Highlights potential connections to scaffold hopping and inverse Quantitative Structure-Activity Relationship (iQSAR).